Hire me for ML engineering and agentic automation
I’m Tyler Renelle, an ML engineer and the host of Machine Learning Guide. Most of my work now is agentic automation: wiring AI agents into a codebase so they plan, write, review, merge and deploy changes, and bring a person only the decisions that need one. I take part-time contract work in both, and my first reply to your email will include my rate and availability.
Machine Learning Guide. I’ve made this podcast since 2017. It’s a free audio course that teaches ML and AI from the ground up: intuition, models, neural networks, the math and the frameworks.
Gnothi, an AI product in production. I built and run Gnothi, live on the web, Google Play and the App Store. It writes and narrates a book for each user through a multi-stage LLM pipeline that plans, researches, drafts and audits every chapter, and it runs causal discovery (PCMCI+) over the data users track.
Agent teams that ship code. Each task gets agents with separate jobs on Claude Code: one plans and coordinates, one writes the code, a fresh one reviews the final diff, and a dedicated one writes any prompt changes. They take the task from an isolated branch through tests, review, merge and deploy, then verify what landed.
Nightly unattended maintenance. Every night an agent run reads search and analytics data, ranks what is worth doing, and does it: content and product-data updates, security and dependency checks, and A/B experiments it starts and evaluates itself. It merges its reviewed work and reports what changed.
Agents that find their own work. The agents watch analytics, search data and production error logs. An obvious fix gets made, bigger work gets queued for a later run, and a judgment call reaches me as a single-decision ticket with a recommendation attached.
AWS infrastructure as code. Everything runs on AWS, defined with SST and Pulumi: serverless APIs on Lambda, Aurora PostgreSQL with vector search, Spot batch compute for audio, and a cookieless analytics pipeline queried with Athena.