Capstone Project Overview

A technoeconomic analysis of bypassing a naphtha hydrotreater for reformate production.

An Electric Arc Furnace (EAF) reduces CO₂ emissions in steelmaking by melting scrap with electricity instead of coal and coke, but significant energy is lost through off-gas. This project develops a heat-recovery system to capture sensible heat from off-gas, improving process and plant efficiency while reducing overall energy losses.

McMaster University relies on a closed-loop chilled water system to cool most campus buildings, but rising temperatures and campus growth are straining its capacity. This project aims to improve the heat transfer efficiency of the existing chilled water plant without expanding its footprint, helping the university continue to grow while maintaining comfortable spaces.

Atlantic Packaging aims to improve reliability and sustainability by ensuring effluent streams meet municipal regulations. This project optimizes the effluent treatment process using new and existing equipment to recommend an engineering solution that mitigates and regulates hydrogen sulfide and methane levels before discharge to the city.

Wastewater plant startup can face issues such as poor effluent quality, foaming, bulking, and membrane fouling in MBR systems. Seeding with biomass is often costly, so controlled loading is used to stabilize biology. This project explores integrating SBR-style operation with Fibracast membranes to improve startup performance and long-term foam control.

A compostable, high‑performance barrier coating for paper and molded‑fiber food packaging is developed using a modified polysaccharide system with bio‑additives. The project optimizes formulation and coating processes, validates water, grease, and oxygen resistance, and ensures compostability, feasibility, and strong market positioning as a zero‑waste alternative.

A FEL1 project designing a lithium refinery that converts LiCl eluate into 25,000 t/yr of battery‑grade Li₂CO₃. The goal is a technically feasible, safe, and sustainable flowsheet that meets purity and yield targets, defines the viable feed purchase price, and supports an economically and environmentally responsible plant.

The medical industry generates 150 million tons of medical waste annually, with 25% of that coming from packaging. We are presenting an evaluation of biodegradable packaging options that reduce waste and allow for critical sterilization procedures.

This project evaluates the feasibility of operating a membrane bioreactor system in batch mode during startup. Temporary batch operation is used to promote controlled biomass development before continuous operation.

Molten Ridge Suites is a proposed 10‑storey mixed‑use building in Hinton, Alberta, serving as both a hotel and an emergency shelter for wildfire evacuees. It provides adaptable, affordable, and sustainable housing, with dining and recreation on the lower floors and upper‑level suites that transition between tourist use and evacuee accommodation, supporting community resilience year‑round.

The City Centre Gardens project proposes the redevelopment of the existing abandoned City Centre Mall property located at 77 James Street North in Hamilton, Ontario, into a five-storey mixed-use building that integrates commercial, community, and housing uses. The design includes a publicly accessible outdoor green space designed to encourage social interaction, recreation, and an active lifestyle

The Hoverferry Terminal at Pier 8 in Hamilton, Ontario, is a proposed waterfront building designed to support a new hovercraft transportation system across Hamilton Harbour. The goal of this project is to create a functional, sustainable, and visually appealing terminal that fits within the ongoing redevelopment of the Pier 8 area.

The Victoria Avenue Community Hub is a multi-use building focusing on improving the social determinants of health in Ward 3 of Hamilton, Ontario. It is the first building of its kind to combine affordable housing, medical care, childcare and food access under one roof in an effort to increase life expectancy in this area.

The Paul Dwyer Centre for Learning and Community is a hybrid secondary school and community hub in North Oshawa. Designed to address rising enrollment while integrating shared community spaces, it serves as a scalable, flexible model for future educational infrastructure across Ontario, emphasizing efficiency and long-term community benefit.

Westinghouse Commons is a ten‑storey mixed‑use, mixed‑income development in downtown Hamilton addressing Ontario’s affordable housing shortage. It includes 203 one‑ and two‑bedroom units at both affordable and market rates, along with a large main floor featuring ground‑level units and a modern fitness facility for residents and the surrounding community.

Sterling Consultants Inc. is designing a five-storey vertical greenhouse located in Hamilton, Ontario, Canada to address the rising food costs and increased volume in food banks throughout the city. The complex will include a packaging and distribution area, offices and conference rooms with multiple floors for produce to grow. The building will incorporate multiple engineering disciplines.

The Willow Point Transitional Home is a five‑storey mixed‑use development in Barrie providing safe, supportive housing for women and children leaving crisis. It includes ground‑floor commercial space, shared community areas on the second floor, and upper‑level dorms and apartments for varied family sizes, fostering stability, empowerment, and inclusive grow.

Westinghouse Commons is a ten‑storey mixed‑use redevelopment at Bronte Rd. and Hwy 407 responding to rising commuter demand. Parking shortages, constrained bus circulation, and unsafe pedestrian–vehicle interactions require a redesigned layout. The proposal adds a multi‑storey parking garage and elevated bus deck to expand capacity, separate movements, and improve safety.

Located at the intersection of Weston Road and St. Clair Avenue in Toronto, the proposed Toronto Weston High School will be five stories. The project addresses the rise in nearby residential development and capacity concerns in local schools.

6ix Stadium is a proposed facility in Toronto with a seating capacity of 65,000, developed to support the 2026 World Cup and the city’s growing demand for major events. The project integrates sustainable, innovative, and intelligent infrastructure as well as transportation network improvements to create lasting economic and social benefits.

The Northern Ontario Fire Operations, Training, and Airfield Facility (NOFOTAF) is a centralized training and resource hub in Thunder Bay, designed to address growing wildfire concerns in Ontario. The facility includes everything needed to train firefighters and combat wildfires, including physical training grounds, educational spaces, offices, airfield facilities, firefighter accommodations

The Wazhushk Onigum Nation has in recent years expressed the need to strengthen its local economy and enhance infrastructure through developments that promote tourism, employment and housing availability. The project proposes a hotel featuring a casino and residential units and meets these goals by creating a space that supports economic growth while maintaining a deep respect for the environment

Larch Commons is a six-storey mixed-use mass timber building located at Larch Street and West Broadway that delivers an environmentally responsible and cost-effective solution to residential apartment-style housing, immersed in Vancouver’s Kitsilano neighbourhood. Its mass timber structure, rooftop green spaces, and energy-efficient features provide innovative and sustainable elements.

The purpose of this project is to address the shortage of long-term care in Oakville City hospitals. To address this social problem, the team proposed a five-storey LTC residence right next to the Oakville Trafalgar Memorial Hospital. This facility will provide services to seniors, people with disabilities, and those who need postoperative rehabilitation.

A proposed 16,200 m² expansion building to accommodate the overcapacity of inmates in Maplehurst Correctional Complex in Milton, Ontario.

The proposed project involves designing a new GO station in the city of Grimsby as an additional stop on the Lakeshore West Line. The purpose of this GO station is to provide improved connectivity from Niagara to Hamilton as a solution to alleviate the increased traffic along the QEW corridor

Delivering of quality and economical housing in the North York region through applying innovative techniques and principles to a mid-rise apartment design. Approach of this goal is framed through building upon the ongoing Transit-Oriented Communities and Modular Housing Initiatives by City of Toronto.

Canada’s housing crisis has left many unable to access affordable homes. This project proposes a six‑storey modular condominium at 1460 Victoria Park Ave. in Toronto, using off‑site modular construction to cut costs and timelines. Built on a 0.59‑acre lot, it will provide one‑, two‑, and three‑bedroom units for young professionals and new families.

LitShare is a social book-sharing platform that empowers readers to organize their personal libraries, create curated booklists, and discover their next favourite read through personalized recommendations. By bridging the gap between social networking and literary discovery, it transforms the reading experience into a collaborative community journey.

Dominion Command is an online multiplayer tower defence video game. It involves two players who each must balance defending their base with sending units to attack the other player’s base, the first to destroy the other’s base wins.

Pocket AI helps farmers log herd health events hands‑free. Instead of stopping to type, farmers speak commands like “Cow 5333 is limping” or “Create a task for Cow 4340 before 3pm,” which are converted into structured records. Voice dictation, API integration, and a review dashboard streamline in‑barn updates, reduce paperwork, and support multilingual farm workers.

This project is part of an ongoing research initiative aimed at designing accessible communication technologies to support social participation among older adults. It seeks to enhance engagement, satisfaction, and accessibility through the integration of mixed reality and artificial intelligence.

This capstone project turns AI-reviewer proof-of-concept into a secure production web app. It returns a preliminary decision (Reject/Not Reject) and one reviewer-style comment; editors can edit the comment, give feedback. The app includes role-based access, file upload, automated AI review, and integrity checks (plagiarism and AI-text detection) with human-in-the-loop overrides for transparency

Docnition AI converges siloed enterprise content (PDFs, wikis, decks, tickets) into a single, role‑aware answer engine that understands a company’s internal logic and indexes at the sub‑document level. Every response is traceable to the exact page and paragraph, and teams get instant, auditable answers with quick actions like summaries and emails instead of hours of hunting across tools.

LM Compass is a web-based platform designed to evaluate and rank the quality of responses to queries from multiple Large and Small Language Models (LLMs/SLMs) using a scalable LLM-as-a-judge cross-evaluation system. By integrating user customizable judging rubrics and a variety of evaluation methods, we’re building an open and unbiased framework to determine which model is best suited.

Geckode is a web-based block coding platform designed to make programming intuitive for beginners while providing powerful tools for experienced developers. Unique features such as a true physics engine, real-time collaboration, and new blocks that condense complex game logic into a single block give Geckode significant advantages over platforms like Scratch and MakeCode.

Kollec is a full stack application providing collectible management and inter-user trading in a single solution, powered by a rapid identification AI/ML algorithm. It allows hobbyist to connect with the community through automated trading systems and profile sharing.

Mac-FAST is a tool for practice testing, which is an effective way to study and promote long-term retention of material (Dunlosky, 2013), in addition to being our students’ most preferred study strategy (Greenberg, unpublished results). Mac-FAST will allow for practice testing in introductory chemistry and kinesiology, based on a sizable archive of our prior multiple-choice tests and exams.

Battery SOC estimation relies on advanced algorithms and standardized testing to compare performance across modern EV scenarios. Our cloud platform accelerates R&D by letting users upload and benchmark SOC models at scale, featuring accounts, leaderboards, visualizations, and model comparison tools.

The Slicc project simplifies group expense management by combining event planning, expense splitting through OCR and voice recognition, and payment coordination in a single application. Key features include automatic receipt scanning, voice-based bill splitting, payment tracking by expense creators, configurable tax/tip handling, group-based event organization, and flexible payment integration.

A free and open-source, web-based inventory tracking system designed for non-profits. Our platform streamlines inventory management through barcode scanning, manual entry and bulk uploads, while providing real-time dashboards and machine-learning-based demand forecasting to help organizations anticipate needs, reduce waste and prioritize procurement or donation requests.

Languine is a language vocabulary revision application designed to aid users in the process of developing flashcards through translation, image, and example generation.

APEX is an AI-powered virtual patient simulator that enables medical trainees to practice empathetic communication through realistic, multimodal patient interactions involving text, audio, and emotion-aware responses. The platform provides automated, research-aligned feedback on empathy and clinical communication while offering a scalable, privacy-conscious tool for training and evaluation.

WISPR is a tool that aims to predict the spread of wildfires with sufficient accuracy. Our hope is that by employing modern machine learning approaches, we may produce insights that surpass the accuracy of existing mechanistic models.

Leopard is a web based IDE platform that allows students to complete assessments and instructors to manage assignments. It provides standard code environments for students to code with, a granular replay viewer of the students actions, and a UI to manage all their assignments and classes.

Veritas is a Chrome extension that helps users identify potential bias and misinformation within online text. It analyzes highlighted passages to extract claims, retrieve relevant evidence, and present insights that support more informed reading for users.

Concentric and Archimedean chess are variants with brand-new boards based on a series of concentric circles or Archimedean spirals, with a new piece, to unlock challenges for old and new players of chess alike. Our team has created a web application to enable anyone to play these variants with friends or match with players online!

Cow behaviour is a key indicator of health, but constant human monitoring is impractical. “Cow-puter Vision” uses computer vision and machine learning to track behaviour from video streams, with future potential for identifying individual cows. By automating herd monitoring, it gives farmers actionable insights to improve management and productivity.

Peer.io is a mobile-based application aimed at helping students and young professionals find projects to work on and potential collaborators.

Real-Time Sign Language (RTSL) is an accessible web application relying on machine learning to convert American Sign Language (ASL) gestures to written English, helping bridge the gap between ASL users and non-ASL users.

A centralized event management platform for MES to replace manual processes and external tools.

This project builds a multiplayer platform where children complete tasks in a controlled virtual environment while researchers log behaviour and communication. It includes mini‑games, a researcher dashboard, secure data storage, authentication, and an AI participant trained via reinforcement learning to simulate different interaction styles

CrewForge is a task-agnostic multi-agent LLM system that allows dynamic creation, pruning, and reconfiguration of agent teams based on given task and performance requirements. A Meta Agent oversees this process, allowing for mid-task reflection and adaptation (such as spawning, retiring, or reconfiguring agents) during execution.

A smart fitting room system that bridges physical retail with digital experiences. Customers scan items in fitting rooms, request different sizes/colors from staff, and receive personalized recommendations through a kiosk interface.

IntelliJam is an adaptive study tool that ingests course materials (slides, PDFs, videos, notes) to build a directed concept map of key ideas and prerequisite relationships. It then generates cited practice questions grounded in the source content and updates a per-concept mastery model to decide what to quiz next.

ARDEN lets writers build branching narratives by describing plot points and characters, while the AI generates the graph and tracks lore in a structured database. It supports large stories with persistent context, exports to Unity, and extends Microsoft’s GENEVA research by adding persistence, incremental editing, and engine integration.

Growin is a Windows productivity app that helps users build focus habits through a “plan → focus → feedback” loop. It supports timed focus sessions, monitors the active foreground app, enforces whitelist rules to reduce distractions, and records local history/usage for review, with an extensible design for rewards/pet system, mini-games.

macSocial is a McMaster-exclusive social platform that enables students to discover events, join clubs, and engage through a unified social feed. The platform supports real-time interactions, event announcements, and community engagement using a scalable Firebase-backed architecture.

An automated wildfire‑detection system uses a solar‑powered BLE 5.0 sensor mesh to monitor heat and CO₂, paired with an AI‑enabled drone for rapid visual confirmation. Fusing real‑time environmental data with autonomous imaging aims to cut emergency response times by 20% and sharply reduce false positives.

A smart and modular solution that captures force and motion data from paddle-mounted sensors and transforms it into clear, actionable insights. By combining embedded hardware with an intuitive mobile app, it helps teams improve synchronization, optimize technique, and gain a competitive edge.

This project develops a low-cost acoustic camera that combines multi-channel audio sensing with computer vision to localize and interpret sound sources in real time. By fusing beamforming-based acoustic heatmaps with machine-learning-driven object and sound classification, the system identifies what is producing sound and where it is located within a scene.

PinPoint is a low-cost automated PCB prober that integrates precision XY motion control, computer vision, assisted feature detection, and automated signal acquisition. It enables fast, repeatable, sub-millimeter PCB probing for efficient electrical testing and debugging.

We develop a custom hardware and software solution for wireless power transfer and communication over the same 5.8 GHz carrier, eliminating the need for batteries or separate communication links in low-power endpoints.

SixDot aims to assist people with Usher Syndrome, a person with Usher Syndrome faces hearing loss and progressive vision loss. SixDot is a portable electronic/mechanical braille system that dynamically converts text into braille in real time.

WattWalk is a capstone project that converts footsteps (or movement/pressure on a walkway) into interactive LED lighting.

PointNemo is a chair-based, real-time active noise control system that uses adaptive digital signal processing to reduce unwanted environmental noise and improve listening comfort. The project targets low-frequency and tonal noise commonly found in everyday indoor settings, with an emphasis on practical, deployable performance.

A robotic item retrieval and order preparation system with an inventory management system to speed up the shopping experience.

Tick Tock Go is a portable version of commercial timegraphers used to measure the accuracy and precision of mechanical watches.

An anti-theft bike lock that pairs with a mobile app to let you remotely arm/disarm it, while monitoring for tampering and triggering a loud alarm plus instant phone alerts. If the bike is moved, built-in GPS tracking shares location so you can quickly recover it.

This project is a portable, semi-automated tire lug nut fastening system that uses a vision-based detection module to locate lug nuts, a CoreXY motion platform to position a motorized tool, and controlled torque application to accurately tighten or loosen lug nuts while improving consistency, efficiency, and safety in automotive service environments.

A wheelchair-mounted sensing and training system that uses real-time feedback to support skill development, enhance independent mobility, and improve safety for manual wheelchair users.

Reduced visibility causes many traffic accidents in Canada, with fog linked to dozens of fatalities each year. This project develops a HUD that shows real‑time positions of nearby vehicles using LiDAR, radar, and cameras fused on a Jetson Nano. A reflective windshield overlay displays the data, with design limited by budget, latency, and weather performance.

FITT (Fitting-room Intelligent Tracking & Transport) is an autonomous service robot that uses AI, sensors, and embedded control to monitor fitting room occupancy and cleanliness, guide customers to available rooms, and deliver alternate clothing sizes on demand. The system reduces wait times, improves accessibility, and enhances retail efficiency through contactless, real-time management

Watchtower is a modular aftermarket automotive security system that brings ‘Tesla Sentry Mode’ capabilities to any car. By utilizing a Jetson Nano and a network of embedded sensors, Watchtower offers edge AI and 360° sensor fusion capabilities, delivering intelligent threat detection and real-time remote alerts while remaining cost-effective and maintaining ultra-low power consumption.

Our project applies principles of electrical and computer engineering to enhance communication accessibility for hard-of-hearing and deaf students in academic settings. We are designing an affordable, wearable device that converts live speech into readable text and displays captions to the user in real time.

An end-to-end AI-accelerated system that parses sustainability data from corporate reports into structured, reliable metrics for sustainability and advisory firms. The system significantly reduces reporting time while improving ESG data quality for financial analysis.

SafeLink AI is an AI-powered hazard detection and PPE compliance system designed to improve workplace safety in high-risk environments. By combining camera-based detection with wearable alerts, the system proactively warns workers of unsafe conditions to improve safety and reduce preventable injuries in industrial environments.

Hybrid low-voltage smart grid PCB controller which allows for control of bidirectional power flow using smart algorithms.

The Sustaina‑Grid is a hybrid home energy system combining solar, wind, battery storage, and smart load management to cut grid use during peak hours. Its rotating solar‑wind design boosts renewable capture, while automated controls and real‑time monitoring improve peak‑shaving, reliability, and system transparency.

A portable 360° bicycle awareness device uses four cameras and a Jetson Orin Nano to reconstruct nearby obstacles into a unified 360° occupancy grid, detecting and tracking road users within a 30-metre radius with low latency. It mounts to personal or bike-share bikes, runs for at least two hours on battery, and shows hazards in a simplified smartphone view with encrypted data.

Portable sEMG muscle sensor which can be worn while exercising which transmits muscle fiber recruitment measurements during exercise to our mobile app for insight and analysis.

An automated wound‑tracking app lets patients and clinicians monitor healing remotely. Smartphone images are processed with computer vision to estimate wound size, inflammation, and healing progress, displayed in a simple dashboard. Streak tracking and reminders support consistent updates, and strong privacy measures protect sensitive health data.

A two‑way translation glove supports deaf and hard‑of‑hearing communication by converting ASL gestures into speech and turning spoken responses into text through a mobile app. With gesture recognition and environmental sound alerts, it enhances independence, safety, and social inclusion in both English and French.

The goal of Sonolite is to design and prototype an end-to-end ultrasound imaging system capable of core B-mode imaging functionality while prioritizing modularity and minimization. A proof-of-concept that addresses the need for a low cost, portable ultrasound device while aiming for defined image processing.

A small wearable device which monitors and tracks the amount of UV exposure a person would get everyday. Using advanced algorithms and UV sensor to measure UV levels and give immediate feedback using LED indicators. This will help determine if the user has had insufficient, partial or sufficient exposure. Data is sent to a web dashboard where short and long term trends can be viewed.

Ivy is an all‑in‑one web platform that replaces the many fragmented tools students use. It unifies AI‑powered study features and career prep in one hub, automating support, scheduling deadlines, generating summaries, quizzes, flashcards (including from MP4s), and offering resume feedback, cover‑letter generation, and a Pomodoro timer for a streamlined academic workflow.

An augmented reality–enabled Wi-Fi signal strength scanner that provides real-time visualization and generates downloadable heatmap floor plans for network analysis.

A targetable and directional Wi-Fi antenna for the purpose of creating a detailed heatmap to detect signal sources.

We have created a device that monitors loads’ power consumption and can automatically improve their efficiency by correcting for wasted power. The system consists of a microcontroller, sensing circuitry, and switchable capacitor banks to bring inductive loads closer to unity power factor.

An AI‑powered tennis analytics system uses a single 1080p60 outdoor camera and an STM32 edge unit with rain and wind sensors, activated by QR code. A local PC handles ball and player tracking, rally segmentation, and basic scoring. A wearable wristband adds ball speed, foul alerts, shot count, and score, with vibration feedback in progress.

A biomimetic flapping‑wing drone is designed to study efficient low‑speed flight. The project focuses on flight control, onboard sensing, wing actuation, and full system integration, supported by mechanical and aerodynamic design. Performance is validated experimentally using a custom electronically controlled wind tunnel.

The McMaster EcoCAR team currently logs only 4 of the 8 critical vehicle CAN buses, with an optional laptop-based logger adding access to just one more. Our project aims to add a new, in‑vehicle network node that gateways all 8 buses into two consolidated data paths for output to an external device.

A device designed to be installed inside a duct to actively reduce noise, with applicability to other enclosed acoustic systems. It uses a feedforward FxLMS adaptive algorithm, validated through simulation and implemented in real time on an embedded system. The system is deployed inside an HVAC metal duct to evaluate adaptive noise cancellation performance in a realistic environment.

The project implements an autonomous navigation stack for a quadrotor drone capable of operating in GPS-denied indoor environments. By combining onboard vision and inertial sensing with a real-time flight controller, the system estimates the drone’s position and enables stable autonomous flight using open-source robotics software.

Our project aims to extend the flight time of a quadcopter drone by integrating solar power into its energy system. The drone will be equipped with two lightweight, high-efficiency solar panels and a multi-battery configuration to allow in-flight charging. By improving the solar panel voltage to the required battery charging level, the drone can recharge during flight and reduce reliance on ground

A proof‑of‑concept laser‑based wireless power system explores using a dedicated “power drone” to extend mission endurance. A transmitter converts electrical energy into a focused laser beam, while a receiver drone converts the beam back into electrical power and stores it in a battery, demonstrating airborne drone‑to‑drone power transfer.

This interdisciplinary capstone project develops a high-power inverter platform that advances the design of scalable, high-efficiency electric drivetrains. The work targets improvements in next-generation electric mobility and power electronics systems, demonstrated through real-world deployment on the McMaster Formula Electric racecar.

A low-cost, wearable assistive device that provides real-time, audible identification of surrounding objects and their distances, enhancing environmental awareness, safety, and independence for individuals with visual impairments. By translating visual information into spoken feedback, the system aims to help users better navigate their immediate surroundings and improve day-to-day mobility.

An intelligent beverage dispenser that uses computer vision to automatically detect the size of any container intended for beverage consumption and dispense the exact amount of liquid required.

Wireless charger capable of locating and positioning itself under phones placed on a table

F.R.E.D. is a fridge‑inventory and spoilage‑detection system that uses sensors to monitor conditions and image recognition to classify foods and estimate freshness. It tracks all items, predicts how long they’ll last, and sends alerts when something should be used soon, helping reduce waste and simplify meal planning.

Hydro Buddy transforms residential bathtubs into active safety environments by utilizing a multi-sensor array of ultrasonic, acoustic, and motion detectors to distinguish between normal bathing and genuine distress. Upon detecting a confirmed drowning signature, the system autonomously intervenes by triggering audiovisual alarms and engaging a high-speed drainage mechanism to remove the hazard.

PhysioBrain is a prototype rehab system that gamifies physiotherapy using motion tracking and adjustable electromagnet resistance. It shows how sensors, control systems, and real‑time feedback can make exercises engaging. Designed to boost patient compliance and support recovery, it also gives physiotherapists an all‑in‑one tool to prescribe and monitor exercises.

WalkWise is a smart insole system developed to analyze human gait using real-time plantar pressure data. The system integrates force-sensitive resistors (FSRs) embedded across key regions of the foot to capture pressure distribution during walking and standing.

The system acquires neural data via scalp electrodes, employs machine learning algorithms to classify distinct brain states, and visualizes the analysis through an intuitive user interface.

This project develops an autonomous drone docking and battery-swapping system that enables near-continuous operation without human intervention. By automatically managing power, docking, and recharging, the system reduces downtime and extends the usefulness of drones in long-duration applications.

This project focuses on designing and prototyping a sleepiness detection and alert device that uses sensor data combined with machine learning to detect driver drowsiness in real time. When fatigue is identified, the system issues a clear and immediate alert to help prevent accidents and improve road safety.

This project develops an autonomous drone swarm that achieves cooperative flight using real-time inter-drone communication. By integrating multiple sensors and onboard algorithm computation, the system enables drones to maintain formation and adapt to their surroundings.

This project explores a custom wireless battery monitoring system that collects battery data and displays it through a web-based dashboard. The system enables real-time visibility of key battery information for monitoring and electric vehicle application use.

A wearable eye-tracking headset that uses infrared cameras and LEDs to capture and analyze eye movements in real time to control a computer mouse pointer, supporting hands-free interaction and vision-based research applications.

A gesture‑controlled drone uses a wearable glove with IMUs and flex sensors to provide intuitive, real‑time control of flight and object handling. The glove sends processed gestures to the drone’s controller, enabling smooth 3D movement and safe operation. A servo gripper carries small objects, supporting uses like assistive robotics, inspection, and light delivery.

DeskGuard is a smart desk companion that keeps watch over your valuables when you’re not around. It uses computer vision and motion sensing to detect unauthorized access or tampering at personal desks in public spaces like libraries and cafés.

An ergonomic split keyboard with motorized physical adjustability, designed to reduce musculoskeletal strain during prolonged computer use in order to alleviate or prevent certain health problems including repetitive strain injury and carpal tunnel syndrome. The keyboard uses pressure sensors under the keys to automatically make adjustments.

GestuRead is a glove that is used to translate gestures (adapted from American Sign Language (ASL)) into text or text-to-speech (TTS). The glove is able to track hand movement and position for each finger using multiple attached sensors. It then sends the data to a machine learning model which confirms translations based on a trained gestures dataset to aid in communication.

As the field of quantum optics continues to grow, there is an emerging need for low-cost and accessible laboratory equipment. SPICE (Single Photon Incidence Correlation Equipment) is a table-top, modular, Hanbury Brown Twiss interferometer which demonstrates the discrete nature of light through second order correlation measurements using low-cost MPPC single photon detectors.

CAP is a helmet-integrated sensing system focused on impact detection to any helmeted activity.

Remote monitoring plant pot, with a Tamagotchi inspired user interface.

BEAM – Beam-Enabled Acoustic Monitoring – is a low-cost, laser-based microphone system that allows law enforcement and first responders to capture conversations and sounds remotely, without placing themselves in danger.

We are developing a bezel-less large-scale rear projection display using frame-sequential LED illumination and LCD edge blending to create a seamless, high-resolution viewing experience without visible screen borders.

Our project offers a monitoring device for nuclear waste canisters that can pre-emptively track, in real-time, the location and arrival-time of stress crack formation in a single, low-cost, and remote device.

This project develops a maintenance-free power solution for monitoring sensors by harvesting indoor ambient light, eliminating battery replacement and reducing long-term costs. A custom PCB with energy-harvesting power management, adaptive reporting, and data visualization was designed and validated to ensure reliable operation in low-light nuclear facility environments.

Our project develops a radiation dose rate mapping system for the McMaster Nuclear Reactor to survey areas around the reactor core during outages. Cameras and AprilTags determine the position of a submersible via triangulation, while a tethered probe measures dose rate. Position and radiation data are combined to generate a live, readable 3D map for operators.

CTRL‑G(LUCOSE) links a continuous glucose monitor with biomanufacturing systems to provide real‑time glucose data and enable closed‑loop control without costly lab instruments. By removing frequent manual sampling and autonomously optimizing culture conditions, it fills a key gap in low‑cost, real‑time metabolic monitoring for cell cultures.

The project designs and prototypes a low-volume closed-loop blood-flow system for pediatric-relevant hemocompatibility testing, featuring a modular test section with in-line flow and pressure sensing that enables rapid insertion and removal of PDMS and other biomaterials without rebuilding the circuit.

An easy-to-use magnetic actuation system that utilizes three magnetic coils to steer a continuum robot for use in ear, nose, and throat (ENT) surgery. The system is designed to enhance precision, safety, and accessibility in robot-assisted ENT interventions by improving fine motion control in highly constrained anatomy.

A workflow was designed to guide surgeons in recording a reproducible endoscopic endonasal video capable of being reconstructed using 3D reconstruction software, which when coupled with an interactive GUI, would allow healthcare providers to annotate a 3D reconstruction of the endonasal cavity.

Perfusion bioreactors are widely used for monoclonal antibody production, but membrane fouling in ATF systems limits performance and operational lifespan. This project develops a 100× smaller ATF system with live pressure monitoring to model and predict fouling behaviour before critical failure, improving process robustness.

This project is developing a biocompatible nanogel that will contain therapeutic nanoparticles in a larger assembly for stable circulation in the body. Upon the application of high-intensity focused ultrasound, the nanogel will degrade and release the smaller, functionalized nanoparticles for targeted cancer treatment.

Using optical hardware, this project aims to offload the convolutional operation from GPU to free space; enhancing surgical segmentation for real time image guided surgery.

Tourniquets are used by military and emergency personnel to control extreme bleeding of the extremities by applying circumferential pressure. AutoQuet is a semi-automated tourniquet that applies the ideal amount of pressure, which can improve patient outcomes and reduce complications.

Oral diseases are widespread and often worsen when care is delayed due to cost, access barriers, low health literacy, and anxiety, especially in low‑ and middle‑income regions. A dental assistant highlighted the need for an accessible, educational screening tool. The Selfie Dental Assistant app combines image analysis and a symptom questionnaire to offer early screening and preventive guidance.

An automated syringe filling system capable of transferring viscous, bubble-free hydrogels from mixing cups directly into sterile syringes for 3D bioprinting. This device will replace the current manual process that is prone to contamination, and will eliminate a critical manufacturing bottleneck at Tessella Biosciences.

An instrumented platform measures impact forces in slide tackles using a sensorized cleat, a biofidelic lower‑leg surrogate, and a positioning jig. It enables controlled, repeatable testing of force and pressure transfer through protective gear while keeping participants safe. The system improves shin‑guard evaluation and understanding of lower‑leg injury mechanisms.

A physiologically relevant testing device that applies reproducible, in vivo-like cyclic axial loading to lumbar spinal segments with pedicle screws. This device will enable evaluation and optimization of screw designs and surgical trajectories to improve long-term pedicle screw fixation stability.

NeuroLeaf is a brain-computer interface that mechanically turns the pages of a physical book. The system determines user intent by decoding EEG signals into forward or backward page-turning commands, allowing for fully hands-free operation and improving quality of life for individuals with upper-limb paralysis.

A switch-adapted guitar for children with limited hand motor function.

Our solution is a spinal compression support harness paired with EMS textile electrodes that activate deep postural muscles. The harness can be worn during daily activities or exercise to restore spinal loading, while the EMS is used during workouts to enhance muscle engagement and reduce microgravity-induced atrophy.

Developing a system to capture a range of images of a patient’s head, compute a photogrammetric mesh and coregister this surface geometry with the patient’s MRI scans in order to produce a navigable anatomical model for use in surgical planning.

An automated modular robotic spotting system that integrates computer vision with electrical components and a mechanical backbone to autonomously detect, track, and assist the user, improving safety and consistency over manual spotting.

Enhancing the contact lens space by designing the framework for reducing synthetic polymers found within contact lens. Focusing on introducing both a natural polymer backbone and a natural cross-linking material that will have similar biocompatibility, mechanical strength and transparency of a traditional synthetic polymer lens.

Skintelligence is a wearable, wireless system for remote psychophysiological monitoring, enabling researchers to assess emotional responses in natural, home-based environments. Integrating multiple sensors allows for simultaneous monitoring, reaching diverse groups, and expanding biometric studies.

This project is an anatomically realistic surrogate model for quantifying contact forces from hockey related impacts to the arm. This serves as a key stepping stone for developing protective equipment to the upper extremities for high-level hockey players.

Pulse-ED rethinks CPR training as a portable and accessible experience by integrating a physical training device with a mobile interface for real-time guidance. The system is designed to equip more bystanders with the confidence and practical skills to respond during cardiac emergencies, while lowering barriers to access CPR education.

Sacrificial bioinks are used in tissue engineering for modelling of vascularized tissues and hollow structures for high-throughput drug testing, providing a more economic and environmental friendly alternative to animal testing. This project has developed a sacrificial bioink that addresses the common limitations of current bioinks.

A real‑time GUI uses intraoperative endoscope video to generate 3D reconstructions during endoscopic endonasal surgery. By combining photogrammetry with spatial localization, it aims to mark surgical margins more accurately, reduce observer variability, and support better oncologic decision‑making and outcomes.

A phased array of acoustic transducers for contactless manipulation of micro-devices. Applications include assembly of micro-robots and guidance of intravenous therapeutics.

We are developing an instrumented shoulder simulator that enables healthcare trainees to practice anterior glenohumeral dislocation reductions using the Kocher maneuver in a safe and realistic environment. By providing real-time sensing and feedback during each attempt, the device aims to accelerate skill acquisition, reduce patient risk, and improve clinical outcomes.

A height-adjustable drop tower capable of safely and accurately releasing objects of varying shapes and sizes in free fall. The design is to be used in conjunction with an anthropomorphic test device (ATD) for evaluating the likelihood and severity of accidental head injury in legal cases.

Teaching Hands to Save Lives is a portable, low-cost CPR training system that integrates a sensor-equipped manikin with a mobile app to provide real-time feedback on compression depth, rate, and hand placement. The design emphasizes cultural inclusivity and accurate anatomical representation to support effective, self-directed CPR training for bystanders.

A semi-autonomous robotic system that controls laparoscope positioning during surgery without the need for an additional set of hands.

Lab-based experimental set-up testing the use of mechanical vibration to extract hip implants from bone cement during hip arthroplasty revision surgeries.

A hands-free board game controlled using EOG signals from vertical eye movement and blinks to enable autonomous participation of activities for individuals with locked-in syndrome.

Hearpoint.ai is developing next-generation, personalized hearing technology that lets users “tune in” to the voice that matters most, even in loud, multi-speaker environments. Our system performs on-device speaker enrollment to generate a unique voice signature, then uses low-latency target speech extraction to amplify the chosen speaker while suppressing other voices and noises.

This capstone project focuses on reducing the time required for forming an abrasion-resistant magnetite layer on the surface of sintered steel parts via a steam treatment oxidation process, with a primary objective of modifying steam treatment parameters at the Johnson Electric Ancaster facility to improve automotive component production efficiency.

Due to the increased load requirements of electric vehicle gears, we aim to determine what carbon content and heat treatment process yields the highest ultimate tensile strength for two powder metal molybdenum steel grades, FL-4405 and FL-4905.

This capstone project develops an experimentally validated post-weld heat treatment (PWHT) framework for high-strength steels to reduce heat-affected zone brittleness. Laboratory furnace trials are correlated with industrial resistance-heating data to ensure transferability to industrial-scale uses.

This project focuses on identifying and evaluating materials and liner systems that can resist corrosion caused by highly acidic fast pyrolysis bio-oil used in carbon dioxide removal processes. Through experimental corrosion testing and life-cycle considerations, the project aims to recommend cost-effective and durable solutions for the safe storage, transport, and sequestration of bio-oil.

The goal of this project is to investigate whether a boriding medium (boron and alumina) supplied by Gosco Valves can produce uniform, adherent TiB/TiB2 layers on Ti-6Al-4V under laboratory constraints. The goal is to achieve a coating 40-70 µm thick, a surface hardness of 1200 HV, and minimal cracking or spallation, while maintaining the alloy microstructure at boriding temperature as low as 925

This project aims to improve the current process employed by Liburdi Turbine Services for removing diffusion-based aluminide coatings doped with platinum from serviced gas turbine blades. The proposed solution incorporates novel techniques in order to enable efficient removal and verification, introduce platinum recovery and minimize the loss of base material.

Theoretical process to valorize steelmaking slag on an industrial scale to reduce waste and environmental harm.

Solderability testing ensures strong, reliable joints by measuring how well solder wets component surfaces. When tests produce false failures, results become unreliable, causing delays and unnecessary e‑waste. Tin Titans aims to identify and eliminate the source of these false failures so solderability testing can remain accurate and trustworthy.

Development of a testing apparatus and associated testing procedures to quantify the ergonomic criteria of grip and dexterity. The results of the testing apparatus are to be used to provide standard measures of the criteria to better inform and protect consumers, with the ultimate goal of providing foundational data to develop industry standards.

Redesign an existing Heat Exchanger System for industry partner, ArcelorMittal Dofasco, through detailed thermal analysis and simulation tools.

This project focuses on the design, fabrication, and analysis of a shell and tube immersion heat exchanger optimized for experimental flow visualization using Particle Image Velocimetry (PIV). A physical heat exchanger and a matching Computational Fluid Dynamics (CFD) model are developed in parallel, allowing experimentally measured flow data to be used to validate and refine CFD predictions.

Create and implement a solution in an open wind tunnel that reduces flow non-uniformity and turbulence intensity. Calibrate and use hot-wire anemometry in a 2D traverse axis system to characterize results and validate implemented solutions

The goal of this project was to improve a jet facility to better facilitate the study of turbulent mixing of stratified jets in a density gradient. Efforts were focused on improving and redesigning the supply system, nozzle, and measurement tank to better achieve our target flow characteristics.

This project involves the design and fabrication of a desktop ball mill apparatus to rotate a variety of different ceramic ball mill jars, each filled with grinding media, for the purposes of grinding material to be used in lab experiments at McMaster.

An FEA model of a Formula Student composite monocoque is built using material properties from tested coupons to predict the chassis torsional behaviour. These predictions are checked through physical testing on the Mac Formula Electric chassis using a custom torsion rig that simulates the maximum loading scenario.

This project aims to create a portable wind turbine that can be used in remote locations to charge electronic devices.

A pneumatically actuated, soft-bodied robot inspired by worm locomotion is proposed for post-earthquake search and rescue operations. The objective is to navigate narrow, unstable rubble environments that are inaccessible or unsafe for human rescuers and rescue dogs. The design employs augers, an anchoring system and pneumatic actuators to achieve flexible, controlled burrowing motion.

This project aims to create an affordable robotic arm with reliable force-controllable joints for real world interaction tasks.

A foldable robotic arm for space applications that stows compactly for launch, then autonomously unfolds using only its shoulder/elbow/wrist joint motors and locks into a zero-backlash, weld-like structure. In the deployed state, it’s designed to match the stiffness of a single rigid boom, enabling self-deployment without the need of a spacewalk.

The goal of this project is to design, build and test a lab-scale vibration-assisted fluidized bed dryer that can be used to understand the physics of potash drying. This will provide a baseline from which the McMaster Institute for Energy Studies (MIES) can develop sustainable improvements to the BHP drying process.

The drying process is one of the most energy intense industrial processes, with the vast majority of conventional drying technologies using fossil fuels to heat air. This project will entail the design, building and testing of a benchmark apparatus for a batch fluidized bed dryer with infrared heating for solid particle drying. This will reduce carbon emission.

This project involves designing, building, and testing a portable measurement system for commercial kitchen appliances such as fryers and dishwashers to quantify recoverable waste heat and estimate daily energy savings from heat reclamation. The system will collect temperature, flow rate, and gas composition data to use in calculating potential energy reuse based on heat transfer and fluid dynamic

McMaster Baja Racing builds a single‑seat off‑road vehicle, including its critical CVT transmission, a complex system prone to failure. Because the CVT is finished before the rest of the car, the team created an offline test fixture using a gear pump and orifice to simulate variable loads, exposing failure modes early and improving reliability before competition.

Our project focuses on the design and development of a mechanical transfer assistive device to safely and independently transfer a user with limited leg mobility from one seated position to another. The device aims to reduce physical strain on caregivers while prioritizing comfort and usability for users.

Our project focuses on the design and development of a mechanical transfer assistive device to safely and independently transfer a user with limited leg mobility from one seated position to another. The device aims to reduce physical strain on caregivers while prioritizing comfort and usability for users.

The main challenge is to create a cost-effective, single-axis dynamometer capable of accurately measuring forces during metal cutting processes. The design must ensure high sensitivity, minimal cross-talk, and structural rigidity while remaining manufacturable with available resources.

Pneumatic end effector developed for lift-assist systems in automotive assembly. The project confronts the longstanding challenge of component handling and end effector versatility in industrial environments.

Downsize a laminar flow device that can produce a coherent jet through flow conditioning and hydraulic flip. The exit flow will be suitable for machining applications.

Design and fabrication of a specialized test rig for the controlled demonstration of metal cutting phenomena, ensuring operational safety and educational effectiveness. The setup facilitates systematic investigation of cutting forces, chip formation, and tool–workpiece interaction.

This project aims to achieve a fully autonomous litter collection robot intended for city park cleaning. The robot can collect litter of different sizes, is safe for public use, and is capable of navigating around obstacles.

An autonomous robot that aids firefighters with the suppression of electric vehicle battery fires.

A mobile autonomous robot engineered for precision weed extraction in row-crop environments. The system utilizes AI object detection for real-time weed classification and a cable-driven robotic claw for mechanical extraction. By integrating an electric actuator with a cable transmission, the design achieves high-torque extraction to cover a wide range of soil conditions, eliminating chemical usage

Design, Build, and Testing of an Automated Fixture to Test Pouch Battery Cells for Electric Vehicle Battery Packs.

The projects main goal is to automate the testing of prismatic cells used by Centre for Mechatronics and Hybrid technologies. The system must be able to test cells of different sizes. The system should be aesthetically pleasing as it will be shown at trade shows. The system must be transportable and durable.

Design, manufacture, and testing of a thermal management solution specific to the McMaster FSAE team’s custom inverter, meeting necessary thermal performance specifications and packaging integration.

This project focuses on the design, construction, and testing of a dedicated pelletizing press. This system transforms irregular biomass materials into uniform, high-density cylindrical pellets to streamline transportation and maximize overall energy utility.

This project involves a compact, rotating mill capable of reducing biomass such as wood chips and oat hulls to uniform sizes of 2 mm grains. This device will be used in lab scale research of sustainable alternative energy sources.

Developing a scale model of an airport baggage sorting system, automatically diverting baggage to one of four assigned outputs.

This project aims to design, manufacture, and implement a functional airbrake system for the McMaster Rocketry Team. The airbrakes will provide controlled deceleration during descent or late-stage ascent, ensuring improved stability and precision in landing or apogee targeting. The system will integrate mechanical deployment mechanisms, sensors, and control systems to achieve autonomous operation.

This project involves the development of a modular benchtop rig that replicates slag foaming behavior in an electric arc furnace. Its purpose is to generate experimental data to support the validation and improvement of computational slag foaming models.

Design, manufacture, and testing of a flexible hydrofoil that can passively deform due to hydrodynamic forces to reduce camber and drag at high speeds, thus increasing efficiency and performance.

This project designs a benchtop-scale rig to model corium during a nuclear core collapse in a CANDU reactor. The rig investigates feasible heating methods and simulant materials to replicate corium decay heat and phase-change behaviour.

Following the completion of a mechanical contracting bid for a waterpark in the first term as part of the Mechanical Contractors Association of America (MCAA) student competition, the team is developing a water reclamation system to reduce the water requirements of the park.

The team worked with the Canadian Remembrance Torch charity to develop an exoskeleton arm support that will be utilized in remembrance events to celebrate with pride, following their motto of Peace, Freedom and Gratitude.

To design and construct a comprehensive cardiac simulator capable of replicating and capturing the dynamic functions of all four chambers of the human heart—right atrium, right ventricle, left atrium, and left ventricle. The simulator will serve as an educational and diagnostic tool.

An automatic pill dispenser designed to help patients manage their medications by organizing dosage, drug type, and scheduled intake times. It streamlines daily routines and reduces the risk of missed or incorrect doses.

Project QUACK is a bipedal social robot designed to facilitate emotional support and engagement across diverse assistive applications and high-stress settings. Our team is developing an interactive robot through the integration of custom electrical and mechanical systems with a dedicated interaction stack, bridging the gap between stable locomotion and meaningful human-robot engagement.

Individuals with visual impairments often face challenges navigating unfamiliar spaces, limiting independence. D.A.R.E. is a compact wearable device that integrates hardware, routing software, and a user interface to determine location and provide navigation through haptic feedback. This discrete system helps users travel through public spaces more safely, confidently, and independently.

An arcade/box style controller series of children modules that can be mechanically connected to each other, allowing gamers to develop their own combination and style of controller. The children modules will be connected to a parent module which will communicate the inputs of the children modules to the game.

Rather than needing to remove a CNC machine tool head to visually inspect for wear, we are developing a solution to wirelessly transmit vibration data from a sensor placed on the tool head. This data will be stored in a larger database, allowing a technician to analyze and determine the current wear of the tool head without needing to remove it from the machine.

Digital twin system to run on the human follower companion. Its primary purpose is to detect, track and follow a designated human by means of sensing and interpreting the surroundings. The physical twin collects and communicates information about its surroundings to the digital twin, which in turn performs computationally expensive tasks to direct the physical twin in its behavior.

This project implements a high-performance vision-based racing system on a 1/10th scale RC car platform (F1TENTH). Using a dedicated camera pipeline, the vehicle tracks track markers to maintain high-speed stability. The architecture is designed for progressive autonomy, evolving from robust lane-keeping to optimal control strategies for minimizing lap times.

The project aims to develop a system of software and hardware components to significantly improve the overall safety, minimize external/environmental risks and perform safe operation of the Unitree Go2 Air Robotic Dog within a constantly changing hospital setting. SNOOPI will be able to detect obstacles and hazards, determine the best course of action for itself, and deploy effectively and safely.

This project addresses the lack of publicly available 3D models of the Niagara Escarpment’s crevice caves by developing a low-cost mapping system designed for dark, confined spaces. The goal is to make high-quality cave mapping achievable without specialized, expensive equipment.

This project is about the system design of a Flexible Robotic Production System. The system is a modular robotic manufacturing platform designed to support flexible production and repeatable operation within a defined work cell. At a higher level, the system is intended to operate within a multi-workbench production environment, where robotic resources are not limited to fixed task roles.

A birdhouse-shaped automated home garden watering turret.

We develop wireless sensor hardware and software which, after harvesting electricity from low-power sources like solar and RF, captures and processes an image through an FPGA accelerated neural network. Featuring micro-power electronics, custom PCBs, Edge-AI, and sea turtles.

This project proposes the design, development, and fabrication of a next-generation wireless follow focus system for professional cinematography. The system’s core innovation is a closed-loop, bi-directional haptic feedback mechanism that creates a true physical connection between the operator and the camera lens.

RIPPLE is a robotic system that is designed to collect small floating trash on the surface of local ponds. It will be remotely controlled by the user and communicate its location back to a User Interface (UI). A camera stream will also be displayed on the UI to allow the user to identify obstacles in the system’s path.

The project improves the swinging motion of a shaft-sinking derrick during material dumping. By adding an updated motor, sensors, and actuators, the system transports material more accurately, safely, and efficiently. A hardware-in-the-loop simulation integrates multiple software tools to test the control system, reducing cycle time, easing operator workload, and improving overall performance.

SnapReport helps organizations generate reports and improve editing workflows. It quickly converts raw information into structured reports while preserving the company’s tone and style. Integrated with a physical device that collects on-site data (e.g., material properties and dimensions), the platform streamlines reporting, manages project information, and improves report accuracy.

Neptune is an autonomous life-saving system that uses real-time computer vision to detect drowning individuals and rapidly deploy a life-saving flotation device. By integrating intelligent software with a precision robotic launcher, the system provides an immediate, reliable intervention for aquatic environments.

This project introduces an automated AI system for detecting lameness in dairy cattle using synchronized cameras and edge computing to monitor cow gait. MI models analyze video in real time to identify early signs of lameness. A review dashboard allows staff feedback, while a sensor-based wiper cleans camera lenses, ensuring reliable, low-maintenance operation and improved herd health.

The project aims to create a low-cost desktop printed circuit board (PCB) manufacturing system for small-scale and rapid prototyping use. The system will accurately place components in the correct orientation on one side of the PCB. It will then reflow the solder to complete the assembly process.

HALO is a wearable system for infants that continuously monitors health and environmental metrics. It integrates multi-sensor hardware, a low-power embedded system, and safe, lightweight housing. Embedded firmware enables real-time data collection and analysis, while a companion app provides dashboards, alerts, and trends. Machine learning analyzes infant crying to identify possible needs.

This project was centered around developing a visionless lane keeping system for a small-scale vehicle, built on the MARC’s mini-mag city platform. By relying on non-visual sensing, the system provides a reliable lane keeping system under low-visibility conditions like snow storms or heavy fog where traditional vision-based sensing would struggle.

20,000+ scattered Buddhist manuscript fragments hide texts we’ve never read. Our desktop app combines PyTorch-powered computer vision with intelligent matching algorithms to transform months of manual research into hours of AI-guided discovery.

MASK (McMaster Admin for Software and Komputing) is a web app that helps McMaster CAS graduate students and admins track program milestones and requirements, submitting updates, assigning tasks, and seeing progress in one place. It centralizes milestone workflows (with assignees, counters/schedules, and completion status) and supports notifications/visibility so everyone stays aligned.

RoCam is an edge-deployed autonomous rocket tracker combining low-latency computer vision, closed-loop gimbal control, and software stabilization to maintain precision target lock throughout flight. It delivers real-time monitoring via HDMI output and a web dashboard for tracking status, stream review, and mission-time decision support.

A powerful event management and survey platform built for the McMaster Engineering Society. Centralizes event hosting, attendee management, modular survey creation, and data analytics into one easy to use platform.

The MES manages reimbursements for 61 clubs but relies on Google Forms and spreadsheets, causing inefficiencies and errors. This project builds a centralized platform for submitting, tracking, and managing expenses, using React/Next.js, AI‑powered OCR for receipts, secure storage, and cloud integration to streamline workflows and improve auditing.

Audio360 is a software system for smart glasses to improve the spatial and environmental awareness of individuals with auditory impairment. The main feature of the smart glasses will bring users more awareness to their surroundings by alerting them of hazardous audio sources in their environment.

Ad Natura is an artistic piece framed as a 2D puzzle platformer game that incorporates slime mould entities as a unique gameplay mechanic. Through interacting with the slime mould with their avatar, the player learns about the slime mould’s queer properties that make it stand out from other living organisms.

AlgoCatan is a deep reinforcement learning agent capable of playing the board game “Settlers of Catan”.

Flow is a keyboard-only text and diagram editor that extends Vim’s philosophy beyond text, making speed of thought the bottleneck for note-taking, rather than tools or mouse input. It introduces a Vim style “shape objects” and a diagram mode, enabling fast, repeatable creation and manipulation of diagrams with precise keystrokes.

AutoVox is a system that provides an interactive interface to visualize, edit, and export voxel data for 3D printing workflows. It automates converting 3D models into voxel-based representations for multi-material 3D printing involving per-voxel magnetization.

In order to increase the efficacy of physical rehabilitation of an injured individual, our team has developed a tool that can provide feedback and corrections for any prescribed exercise.

The MES runs large student events but relies on scattered tools, making registration, waivers, and check‑ins inefficient. This project creates a centralized platform for tickets, signups, accessibility and dietary info, waitlists, photo galleries, admin dashboards, and notifications, improving organization and the attendee experience.

VoiceBridge is an accessibility-focused web interface that enables individuals with dysarthric or atypical speech to interact with browser-based technology more independently. By combining speech recognition with intent confirmation, the system provides reliable hands-free control of web applications, reducing barriers to everyday digital interaction.

This project is about building a behavior tree–based framework for robotic mission planning and simulation. The idea is to give operators an interface where they can design, visualize, and test mission plans as behavior trees. These trees provide a structured way to break down a robot’s actions and decisions, making mission planning easier to understand and more flexible to modify.

Pocket AI is a hands-free, voice-powered application that uses AI dictation and automation to help farmers to efficiently manage and streamline their tasks.

Hitchly is a campus focused ridesharing app for the McMaster University community that matches verified drivers and riders based on schedule and location. It uses McMaster email verification, enables cost-sharing for commutes, and aims to reduce single-occupancy trips and parking congestion on campus.

DomainX replaces spreadsheet‑based software assessments with a traceable, reproducible system for comparing engineering qualities. It streamlines evidence collection from public artifacts, normalizes scores, and generates rankings with sensitivity analysis. Our capstone validates it through an assessment of open‑source neural network libraries.

This project builds a web platform for exploring a large behavioral neuroscience dataset of nearly 20,000 rat trials related to OCD. It enables researchers to search, filter, download, and analyze subsets of data using natural‑language queries, synchronized video playback, and an accessible interface designed for users without programming experience.

Locating specific cows is time‑consuming, requiring farmers to check ear tags manually. “Cow‑puter Vision” uses computer vision and machine learning to identify and track cows by their unique coat patterns from video feeds. Automating recognition streamlines daily tasks, such as separating individual animals from the herd.

Clustr is a decentralized distributed computing platform designed to accelerate complex data processing tasks by parallelizing workloads across a secure network. It offers a unified orchestration layer that provides researchers and developers with real-time monitoring, fault tolerance, and automated resource management.

Reading4All addresses accessibility gaps in STEM education by improving the quality and consistency of alternative (alt) text for technical diagrams. By leveraging AI-assisted captioning and accessibility frameworks, the project enables students with disabilities to independently access critical visual information and promotes more equitable learning experiences in post-secondary courses.

This project builds an AI‑powered assistive platform that makes computers accessible for users who struggle with traditional interfaces. Using speech recognition, natural‑language understanding, and text‑to‑speech, the system lets people control a computer entirely by voice, enabling natural, hands‑free interaction.

On dairy farms, many tasks require manually checking cows’ ear tags. “Cow‑puter Vision” uses machine learning to identify and track individual cows from video feeds by analyzing coat patterns. Automating recognition streamlines daily operations and reduces the need for manual checks.

This project is a web browser game called Crazy 1-0s. It is a variation of the classic card game Crazy 8s with a twist to incorporate counting in different number bases, especially in Base 12 – Dozenal.