I build data pipelines by day and ship side projects somewhere between midnight and regret. Based in Toronto. Mostly awake.
My work sits at the intersection of data engineering, full-stack development, and AI systems. I like things that scale, fail gracefully, and occasionally do what they're supposed to.
Right now I'm focused on a few projects.
Claude Home explores architectural persistence for LLMs. It gives a Claude instance a durable filesystem and scheduled execution so it can re-encounter its own outputs across sessions. Basically, what happens when you give an AI a home directory.
PassFX is a zero-knowledge, local-first terminal password manager built on standard cryptography. It never touches the network. The threat model assumes everything eventually breaks, so the goal is to break gracefully.
Yield makes algorithms visible. It's an interactive visualizer for classic data structures, built for people who learn better when they can watch things move.
Debate Lab pits AI models against each other in structured debate. GPT and Grok argue. Claude moderates. The real question is whether chaos can be well-architected.
I care about hard problems at the intersection of data, systems, and AI. Currently interested in LLM architectures, applied AI research, and finding better ways to explain complex ideas without hand-waving.
Open to full-time roles, collaborations, and technical conversations that start as a "quick question" and quietly become a deep dive.
Languages
Data Engineering & Databases
AI/ML & Data Science
Web Development
Cloud & DevOps
- 🎉 Merged PR #7 in dinesh-git17/claude-runner
- 💪 Opened PR #7 in dinesh-git17/claude-runner
- 🎉 Merged PR #64 in dinesh-git17/claudehome
- 💪 Opened PR #64 in dinesh-git17/claudehome
- 🗣 Commented on #33 in dinesh-git17/techpulse




