Montreal, Canada
Shipping production ML & LLM agents that customers rely on
ML engineer with 6+ years building production ML and LLM agent systems — from RAG, evals, and MCP to biosignal and real-time inference.
About me
With 6+ years in ML engineering, I ship production ML and LLM agent systems for customer-facing workflows — owning quality, evaluation, and reliability across RAG, MCP, and tool use. I won the Intel AI PC Innovation Challenge and led teams building biosignal, audio, and real-time ML systems from research through deployment.
Tech stack
- Python
- PyTorch
- LLMs
- RAG
- MCP
- TensorFlow
- MLflow
- GCP
- AWS
- BigQuery
- FastAPI
- Flutter
- Vue.js
- Docker
- Kafka
My Expertise
-
Data Science & ML
I design and deploy ML systems with PyTorch, TensorFlow, and HuggingFace — from CNNs, GANs, and transformers for biosignal and audio to LLM agents, RAG, and recommendation models on GCP and AWS.
-
MLOps
I build reproducible ML pipelines with MLflow, CI/CD, and LLM-as-a-judge evaluation — automating training, deployment, and quality instrumentation so models stay reliable at scale.
-
Software Engineering
I develop production Python services with FastAPI and Flask, plus cross-platform apps in Flutter and Vue.js — integrating LLM APIs, MCP servers, and real-time inference into user-facing products.
-
Project Management
I lead cross-functional teams to deliver AI on schedule — from the Intel AI PC Innovation Challenge to Google Project Management certification — using Agile to drive measurable outcomes.
