Tory Jarmain
Tory Jarmain
AI Research & Full Stack Development
NVIDIA GTC NeurIPS MICCAI

Biography

Tory advises Fortune 500 companies and startups on AI strategy and implementation, helping them leverage cutting-edge AI technologies for business impact.

He specializes in delivering rapid AI prototypes to elite clients, transforming complex business challenges into working AI solutions within weeks, not months. His ability to architect and deploy production-scale AI systems at unprecedented speed has earned him the trust of industry leaders who need to move fast in competitive markets.

He has been researching AI as a computer scientist since 2012. His work encompasses the application of machine learning algorithms including Deep Learning, Reinforcement Learning, and Large Language Models across domains from medical diagnostics to quantitative trading systems and AI voice and video agents.

Tory founded Triage, a medical device company that developed breakthrough AI-powered dermatology diagnostics and received regulatory approval in 35 countries. His research was funded by NVIDIA and Google, enabling large-scale experimentation at the frontier of applied AI. His contributions have been presented at leading AI conferences NVIDIA GTC, NeurIPS and MICCAI, and he has collaborated with researchers from Stanford, UCLA, MD Anderson Cancer Center, Memorial Sloan Kettering Cancer Center, Vector Institute, and University of Toronto. Tory's peer-reviewed AI research publications include co-authored work with Dr. Ron Moy, MD, former president of the American Academy of Dermatology and Chief of Dermatologic Surgery at UCLA. Previously, Tory was Co-Founder and CTO of technology bootcamp Bitmaker, which was acquired by General Assembly. He graduated from Ivey Business School (HBA 2011) at Western University. Tory's mentors include technology leaders Daniel Kottke of Apple, whose insights have been invaluable to his development.

Building the Future

Inspired by JARVIS from Iron Man, today Tory is building AI agents that collaborate with humans as naturally as another person would. His recent work spans medical research assistants, autonomous workflow systems, voice interaction platforms, and computer-use agents that navigate complex digital environments.

AI is rapidly moving beyond static use cases, like chatbots or diagnostic tools, toward agents that can see, listen, create, and respond in real time. The next five years will be about building systems that feel less like software and more like collaborators — capable of helping us think, create, and make decisions in a way that feels human. Tory aims to make that future a reality, pushing the boundaries of what's possible in human-machine collaboration.

AI Voice & Video Agents

AI Voice & Video Agents

Tory's work in this area has been applied to medical research and education, where voice- and vision-enabled AI can serve as teaching assistants, patient-facing guides, and collaborators in research workflows. These systems also power outbound and inbound sales agents, demonstrating the versatility of embodied AI across different domains.

Key Technologies

  • Computer Vision
  • Speech Synthesis
  • Natural Language Understanding
  • Reinforcement Learning

Applications

  • Medical Research & Education
  • Patient-facing AI Guides
  • Research Workflow Collaboration
  • Teaching Assistants
  • Outbound Sales Agents
  • Inbound Customer Support Agents
Advanced Reasoning Systems

Advanced Reasoning Systems

Building state-of-the-art AI agents from the ground up with chain of thought, tool use, and advanced reasoning capabilities. Tory's work spans from early foundation models to cutting-edge reasoning systems that can think, plan, and execute complex tasks autonomously, including experimentation with recursive self-improvement.

Key Technologies

  • Chain of Thought Reasoning
  • Tool Use & Function Calling
  • Agent Architecture Design
  • Research & Planning Systems
  • Recursive Self-Improvement

Models & Projects

  • GPT-2 to Latest Reasoning Models
  • Autonomous Research Agents
  • Multi-step Reasoning Systems
  • Tool-Integrated AI Agents
  • Self-Improving AI Systems
Generative Media

Generative Media

Exploring how LLMs can drive image and video creation, enabling new forms of expression, simulation, and interaction at the intersection of creativity and machine intelligence.

Key Technologies

  • Image Generation
  • Video Creation
  • Creative AI
  • Multimodal LLMs

Applications

  • Creative Expression
  • Simulation & Modeling
  • Interactive Media
  • Content Creation
Intelligent Research & Workflow Agents

Research & Workflow Agents

Creating sophisticated AI agents that can conduct research, analyze data, and execute complex workflows autonomously. These systems combine reasoning capabilities with practical tool use to solve real-world problems, from research automation to business process optimization.

Key Technologies

  • Research Automation
  • Intelligent Workflow Design
  • Multi-agent Coordination
  • Autonomous Decision Making

Applications

  • Research & Data Analysis
  • Business Process Automation
  • Intelligent Workflow Management
  • Autonomous Problem Solving
  • Outbound and Inbound Customer Support and Sales Agents
Medical AI Research

Medical AI Research

Developing AI systems for medical diagnostics, patient care, and healthcare workflow optimization. Tory's work includes computer vision for medical imaging, natural language processing for understanding clinical records, and AI agents for medical research and education. This includes developing AI for diagnosing dermatological and ophthalmological conditions.

Key Technologies

  • Medical Computer Vision
  • Clinical NLP
  • Medical Workflow Automation
  • Patient Care AI

Applications

  • Medical Imaging Analysis
  • Clinical Documentation
  • Medical Research Automation
  • Healthcare Process Optimization
Financial AI Research

Financial AI Research

Advanced AI systems for financial markets, investment strategies, and quantitative analysis. This research area focuses on developing sophisticated algorithms for market prediction, risk assessment, and automated trading systems.

Key Technologies

  • Quantitative Analysis
  • Market Prediction Models
  • Risk Assessment AI
  • Automated Trading Systems

Applications

  • Investment Strategy Optimization
  • Market Analysis & Prediction
  • Risk Management Systems
  • Portfolio Management AI

Technical Skills

Programming Languages

Python Rust C C++ C# Java Ruby Objective-C JavaScript TypeScript CSS HTML5 SQL XML JSON Go Bash

AI & Machine Learning

PyTorch TensorFlow JAX Neural Networks Reinforcement Learning OpenAI Genetic Algorithms TensorRT-LLM vLLM DeepSpeed Megatron-LM Huggingface LangChain LangGraph CrewAI AutoGen smolagents E2B Browser Use MCP RAG MLX MLX-VLM Local LLMs ExLlamaV3 llama.cpp TensorRT ONNX TorchServe Transformers Stable Baselines OpenAI Gym Deep Q-Learning Model Training Fine-tuning LoRA QLoRA PEFT Inference Model Deployment Data Collection Data Labeling Data Validation Data Quality Text-to-Speech Speech-to-Text Video Generation Text-to-Video Kling AI Multi-modal AI Flux Quantization

MLOps & Infrastructure

Docker Docker Compose Kubernetes Helm Slurm Sidekiq CUDA GPU Optimization Large-scale GPU Clusters AI Inference at Scale Weights & Biases TensorBoard New Relic Datadog

Data Analysis (Healthcare & Finance)

Pandas MATLAB Mathematica R Quantitative Finance Financial modeling Apache Spark BigQuery Bloomberg API S&P Capital IQ Databricks Jupyter

Finance & Accounting

Accounting Finance Financial modeling Quantitative Finance Bloomberg API S&P Capital IQ

Cloud & Infrastructure

AWS S3 AWS EC2 AWS Lambda Google Cloud Google Cloud Vertex Firebase Google Cloud Functions Lambda Labs Azure Hetzner Digital Ocean Vercel Netlify Cloudflare

Vector Databases & RAG

Vector Databases Pinecone Chroma Qdrant Vector Search Embeddings

Security & Compliance

SSL GDPR PIPEDA PHIPA HIPAA ISO 13485 SOC2 EU MDR OAuth Auth0 Security Auditing

APIs & Integration

OpenAI API Gemini API Google Maps API Twilio Mailchimp Resend Stripe Stripe Connect Shopify Zapier n8n Make REST REST APIs GraphQL Pub/Sub messaging PPC Ads (Facebook, Google)

Databases

PostgreSQL Oracle SQLite MySQL Microsoft SQL Server MongoDB NoSQL ElasticSearch PostGIS

Web Development

Node.js Express Flask FastAPI Django Ruby on Rails .NET PHP WordPress React Redux React Native Angular Next.js Vue.js WebRTC WebSockets Microservices Tailwind CSS SASS JSS Flexbox CSS Grid Animation Three.js D3.js Chart.js Responsive Design Progressive Web Apps

Mobile Development

Swift Objective-C Java Android Studio

Computer Vision

OpenCV YOLO MediaPipe Image Processing Object Detection Optical Character Recognition CT DICOM

Virtual Reality / Augmented Reality Development

VR/XR Unity Unreal Engine 5 Meta Quest Meta AI Glasses Frame SDK

Development Tools

Figma GitHub Copilot Replit Cursor VS Code Xcode Postman Project management Systems engineering Requirements gathering Communication skills Test automation Debugging Cloud architecture Asana Notion Trello Git Git Workflows GitHub Actions GitLab CI CI/CD Pipelines Environment Management Webpack Vite npm yarn Lighthouse React DevTools Performance Profiling Auditing Tools Selenium Scrapy nokogiri

Productivity Tools

Excel Word PowerPoint Keynote

Testing & Quality

Test-Driven Development Jest Cypress Playwright Unit Testing Feature Testing Acceptance Testing E2E Testing Integration Testing Testing Frameworks

Hardware Platforms

Apple Silicon NVIDIA GPU Raspberry Pi Cerebras Groq iPhone Android

Performance & Caching

Redis Nginx Load Balancing Caching Multithreading

Healthcare Standards

FHIR HL7 SNOMED CT Healthcare APIs Medical Data Electronic Medical Records Epic Cerner

Operating Systems

Windows 11 Linux macOS iOS Android FreeBSD

Google Sponsorship

Tory's cutting-edge healthcare AI research has been backed by Google Cloud and Jeff Dean, Chief Scientist of Google DeepMind and creator of Gemini

NVIDIA Sponsorship

Received both compute and recognition at NVIDIA GTC for large-scale healthcare AI experimentation

National Research Council Canada Funding

Received research funding from the National Research Council Canada, supporting innovative AI research and development projects

Technology Leadership Mentorship

Received mentorship from technology leaders including Daniel Kottke of Apple, gaining insights into building transformative technology and scaling innovative products

Stanford Research Collaboration

Triage technology assessed by healthcare AI pioneers Dr. Justin Ko, MD and Dr. Roberto Novoa, MD

Memorial Sloan Kettering International Skin Imaging Collaboration

Working group member involved in advancing science of melanoma diagnoses with Dr. Allan C. Halpern, MD, Chief of Dermatology at Memorial Sloan Kettering Cancer Center

Select Publications

JOURNAL ARTICLE

"Artificial Intelligence vs Medical Providers in the Dermoscopic Diagnosis of Melanoma"

2023 • Cutis • 111(5):254–258

Tory Jarmain, et al.

DOI: 10.12788/cutis.0764

JOURNAL ARTICLE

"Diagnosis of Skin Disease in Moderately to Highly Pigmented Skin by Artificial Intelligence"

2023 • Journal of Drugs in Dermatology (JDD) • 22(7):647–652

Tory Jarmain, et al.

DOI: 10.36849/JDD.7581

JOURNAL ARTICLE

"Artificial Intelligence in the Evaluation of Telemedicine Dermatology Patients"

2022 • Journal of Drugs in Dermatology (JDD) • 21(2):191–194

Tory Jarmain, et al.

DOI: 10.36849/jdd.6277

CONFERENCE

"Improving Skin Condition Classification with a Question Answering Model"

2018 • Medical Imaging meets NeurIPS Workshop • NeurIPS 2018

Tory Jarmain, et al.

arXiv:1811.06165

CONFERENCE

"Improving Skin Condition Classification with a Visual Symptom Checker Trained Using Reinforcement Learning"

2019 • MICCAI 2019, Shenzhen, China • Springer, LNCS

Tory Jarmain, et al.

DOI: 10.1007/978-3-030-32251-9_60