Toronto, Canada — Available for projects
I am Alex, a fullstack developer who has also run a small restaurant in Toronto. That combination lets me build software that actually understands how a small business runs day-to-day.
01
I started programming on my own, then went through Le Wagon's fullstack bootcamp, did a three-month internship at Oesia building an internal document-management system, and spent some time teaching Ruby to other students.
I also ran Churreria Calderon, a small family restaurant in Toronto — which gave me a first-hand feel for how small businesses actually operate. Most of my recent work is around AI agents and RAG systems, the kind of tools small businesses now have access to but rarely know how to deploy.
I am comfortable across the stack: Python, JavaScript, TypeScript, Ruby on Rails, PostgreSQL, Docker, and the modern AI tooling layer. I care about software that ships and survives contact with real users.
02
A few recent projects across AI, open source and the web. Repos are public, feel free to look at the code.
A production-ready Retrieval-Augmented Generation chatbot that lets businesses query their internal PDF documents in natural language. Returns answers with page-level source citations and supports real-time document upload.
Role: Architecture, retrieval pipeline, agent design
A drop-in chat widget for small-business websites. It answers customer questions about the business — menu, hours, location, FAQs — grounded only in the business's own information, so it never makes things up. Reusable for any client by editing a single config file.
Role: Design, full build, deployment
Two contributions to fenn, an open-source Python framework for ML/deep-learning workflows with a Retrieval-Augmented Generation module. First I added a .docx (Word) document loader — following the library's existing loader pattern to extract text from paragraphs and tables, with an optional-dependency extra and unit tests. Then I fixed the RAG modules' install instructions, which pointed users to a package and extras that don't exist. Both submitted upstream as pull requests.
Role: Open-source contributor — feature, fix and tests
A recommendation engine built on vector embeddings and a feedback loop that improves results over time. Designed as a reusable backend for content platforms and e-commerce that have outgrown rule-based filters.
Role: End-to-end implementation
A clean, fast website I built for a family-run churreria in Toronto — menu, story and location — and a testbed for how far AI-assisted tooling can take a static marketing site.
Role: Design, build, content
03
I take on a small number of part-time engagements at a time. Open to one-off projects and ongoing collaborations.
Custom chatbots and agents that work with your own documents, data and tools. Built on modern RAG architectures with proper retrieval, evaluation and guardrails.
Fast, modern websites for small businesses, restaurants and service providers. Designed to convert, easy for you to update, and built on stacks you can actually maintain.
I've run a restaurant myself, so I know which problems are worth solving with software and which are not. Reservations, menus, internal automations, customer follow-up.
04
Working on something where AI, automation or a small custom web app would help? Send me a short note, I usually reply within a day.