I am a software engineer based in Canada. I like getting to the heart of a problem before reaching for a tool. Sometimes the right answer is sophisticated, sometimes it is simple, and the judgment to tell the difference is most of the job.
My work spans full-stack product development, applied AI and machine learning, data-intensive systems, and cloud and platform engineering. Real products rarely stay inside one of those categories, so the application, the data, the models, and the infrastructure all have to work together before any of it counts.
- 🔭 Currently — building an end-to-end intelligent systems platform: document retrieval, agent orchestration, analytics, and the pipelines behind them
- 🌱 Exploring — retrieval evaluation, agent reliability, and keeping AI systems honest in production
- 💬 Ask me about — RAG and hybrid search, agentic workflows, explainable ML, or turning a notebook into something people can actually use
- ♟️ Outside engineering — competitive chess, which taught me more about patience and looking past the obvious move than I expected
A representative view of what I have worked with, not a fixed boundary. Most of it was picked up because a project needed it.
Languages
AI / ML & LLM Engineering
Data Science & Machine Learning
Vector Stores & Retrieval
Frontend & Mobile
Backend & APIs
Databases & Data
Cloud, DevOps & Tools
|
Agentic evidence copilot for clinical protocol review. Grounded citations, explicit abstention on unsupported claims. LangGraph · LlamaIndex · Bedrock |
Machine learning app predicting cirrhosis stage through a simple clinical-style web interface. Python · scikit-learn · Vercel |
Explainable risk-scoring workflow that supports business decisions instead of acting like a black box. Python · SHAP · SMOTE |
|
Android app that identifies waste items and guides disposal through an on-device ML workflow. Java · TensorFlow Lite · Firebase |
Text classification across multiple NLP representations, with explainability and class-balance handling. scikit-learn · TF-IDF · Word2Vec |
Fine-tuned BERT benchmarked against deep learning baselines for sentiment classification. PyTorch · BERT · LSTM |
| Role | Organization |
|---|---|
| Software Developer | Dexlabs Inc. |
| Software Engineer, Co-op | NorthCommand Gaming Inc. |
| Teaching & Instructional Assistant | Wilfrid Laurier University |
| Lecturer | Drs. Kiran & Pallavi Patel Global University |
| Full Stack Developer, Intern | CSI Software |
Product development, AI and machine learning, data-intensive systems, and platform work.
Based in Canada, happy with remote teams, and open to relocating.
