Paul McInnis

Full-Stack, Web3, ML

About Me


I am a self-starter who enjoys building things from the ground up, and working with cutting-edge technology.

In my spare time I'm working on my music project Scramble Suit, hiking, bicycling, or soldering.

I am always on the lookout for new and exciting opportunities - if you think I would be a good fit, feel free to drop me a line.

Experience


Hypotenuse Labs

Hypotenuse Labs

Senior Software Developer

  • Delivering web2&3 projects for our hype clients as a full-service software consultancy
  • Leading agile, full-stack web teams of 3-10 people in size
  • Roadmapping customer requirements and delivering on-time
  • Integrating web3 smart contracts into web2 frontends
  • Developing and improving machine learning services
  • Launched ambient finance, a cutting-edge decentralized exchange.
Web2 Web3 DEX Angular React Typescript Javascript PostgreSQL Heroku Machine Learning Metamask AWS SSO MongoDB ERC-20 ERC-721 ERC-1155
Darwin AI

DarwinAI

Senior Platform Developer

  • Development in support of DarwinAI’s Explainable AI (XAI), a cutting-edge quantitative explainability tool
  • Translation of research core IP into product-quality micro-services
  • Delivered highly-performant models to enterprise customers in different verticals
  • Deep experience with common machine learning frameworks
  • Participated in COVID-Net initiative, a tool for identifying and quantifying COVID-19 infections within a chest X-Ray
  • Experience quantizing models for embedded deployment, including TPU accelerators.
Explainability TensorFlow Python Keras PyTorch SQL MPI CUDA TensorFlowLite REST Coral
RaveDJ

Rave Inc

Software Developer (AI)

  • Development in support of RaveDJ, an A.I. DJ web-app
  • Development of novel Music Information Retrieval (MIR) IP for DJ mixing backend
  • Development of novel dataset including training of annotators and pipeline
  • Data science driven research and development in support of novel IP
  • Deployment of AI micro-service for live product
TensorFlow Python Kubernetes C MIR Dev-Ops Data Science Google Cloud
UWAFT

UWAFT

Engineering Manager

  • Responsible for Innovation and Electrical sub-teams
  • Advanced Vehicle Technology Competition (AVTC) for the conversion of a Camaro into a hybrid sports car
  • Engineering work in support of novel powertrain design, controls and simulation
  • Development of hardware and software platform for training object detection and tracking algorithms with Waterloo’s autonomous vehicle research group
  • Design and implementation of all high and low-voltage subsystems including fusing
  • Responsible for numerous high-pressure, techical public speaking presentations
CANBus FEM NX Altium Simulation HV AVTC Object Tracking

Projects


JobFunnel

COVID-Net

Veta

EcoCAR

Frost-E-Bike

Synthesizer

Education


University of Waterloo Ontario

Ba.Sc Mechanical and Mechatronics Engineering ( - )

  • Co-op program
  • University of Waterloo Merit Scholarship
  • Undergraduate Student Research Award, Natural Sciences and Engineering Research Council of Canada (NSERC)
  • Final-year project: AWD, winterized, E-Bike (Frost-E-Bike)

Ma.Sc Mechanical and Mechatronics Engineering ( - )

Publications


- Fibrosis-Net: A Tailored Deep Convolutional Neural Network Design for Prediction of Pulmonary Fibrosis Progression From Chest CT Images

Alexander Wong, Jack Lu, Adam Dorfman, Paul McInnis, Mahmoud Famouri, Daniel Manary, James Ren Hou Lee, and Michael Lynch.

- Inadequacy of Current Pediatric Epinephrine Autoinjector Needle Length for Use in Infants and Toddlers

Harold Kim, Chitra Dinakar, Paul McInnis, Dan Rudin, Xavier Benain, William Daley, Elke Platz.

- Comparative Safety Risk and the Use of Repurposed EV Batteries for Stationary Energy Storage

John Catton, Sean B. Walker, Paul McInnis, Michael Fowler, Roydon Fraser, Steven B. Young, Ben Gaffney.

- System and Device for Management of Medication Delivery Devices

Michael Fisher, Alexander Leyn, Richard Edwards, Steven Orzel, Erik Helge Borg, Paul McInnis, Cristian Cherler, Heinz Wolter, John Chinnick, Jacob Stahlbaum.

Contact