Eric Le I build AI agents and developer tools.
I'm Eric, an AI engineer in Gothenburg. I studied physics at Chalmers, got hooked on AI agents during my master's thesis at Volvo, and now spend my days building AI tools and agents, mostly ones I needed myself first.
Outside of work I do calisthenics, play tennis, and travel whenever I can.
Where I've worked and studied
dec 2025 to present
sep to dec 2025
learning react and typescript
Worked through The Odin Project’s full-stack JavaScript path, then started Stackr in TypeScript that December.
jan 2024 to jun 2025
volvo technology, gothenburg
Built an agent that let test engineers get plots and figures from their truck test data by asking in plain language, from my master’s thesis to the team’s real data. Presented it at Generative AI for Automotive USA 2024, an industry conference in Ann Arbor.
jun to aug 2023
internship at bac ha software, hanoi
Reproduced published GAN architectures in TensorFlow for image generation and image-to-image translation, and trained them on GPU.
sep 2022 to jun 2024
msc in complex adaptive systems, chalmers
How simple agents and their environment produce behavior nobody designed.
sep 2019 to jun 2022
bsc in engineering physics, chalmers
Where the modelling habits came from, before any of this was software.
Every project I had kept its own rules for its AI agents, and they drifted apart. canon holds a single copy of them now, synced into every repository, and it runs each change through the same loop: planned by an orchestrator, built by workers in parallel, checked by an independent reviewer. Every repository I have runs on it. On npm as @erclx/canon, with a live build at canon.erclx.dev.
I wanted to know whether retrieval still helped when a model could read the whole EU AI Act. An agent that maps a described system to the articles that apply, cited to the text. Reading the whole Act found more, and retrieval cost about a seventh as much and showed exactly which text it used.
Staging files by hand for every prompt was the slowest part of every planning chat. VS Code extension that stages files across a workspace into one block of LLM context. Installed from the VS Code Marketplace and Open VSX.
I kept retyping the same prompts across three different chat windows. Chrome extension that saves prompts and drops them into Claude, Gemini, and ChatGPT from the chat input. Live on the Chrome Web Store.
Open to work
