Skip to content
View tommulkins's full-sized avatar
😎
Having fun & making good stuff
😎
Having fun & making good stuff

Block or report tommulkins

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
tommulkins/README.md

Hey, I'm Tom 👋

Senior software engineer with 25+ years of production experience, building agentic AI systems and the tooling around them. I run Hermes Agent by Nous Research daily as my dev environment, alongside Cursor IDE, and most of my work is in Python and TypeScript.

cat typing

What I'm into right now

  • LLM agents and orchestration (LangGraph, MCP, OpenAI SDK)
  • Evaluation systems that actually catch regressions (Braintrust, rubrics, failure mode analysis)
  • Local model workflows and frontier model handoffs
  • Boring, durable engineering practices: guardrails, code quality, life-cycle discipline
  • Tools that turn "I wonder if..." into a working prototype by lunch

The 25-year story

Started on LAMP because that's what the internet ran on. Spent years in fintech and payments, building systems where a bug means someone's money is in the wrong place. Moved into product engineering for sales and revenue organizations, where the job is turning a product vision into something the team can actually ship and iterate on. Stuck around through the JS/TS era, Cloudflare, GCP, AWS, and a lot of distributed systems.

Most of my career has been helping people go from "we have an idea" to "it's live and customers love it."

Stack

Python, TypeScript, Go, SQL, React, Next.js, Node.js, LangGraph, MCP, OpenAI SDK, Cursor, local and frontier LLMs, Braintrust, Cloudflare, Docker, AWS, GCP. Elixir-curious.

Domain

PCI DSS Level 1, SOC 2, HIPAA. I've built systems that handle payment processing, health records, and regulated data.

Resume

Full resume

Find me

Pinned Loading

  1. hermes-skill-project-kickoff hermes-skill-project-kickoff Public

    Hermes Agent skill: orchestrate the first 15 minutes of a new project — capture the goal, scaffold a wiki page, break into sprints, wire GitHub checkpoints, set up /new recovery. Adapted from sovth…

    2

  2. prod-evals-cookbook prod-evals-cookbook Public

    Forked from Gauntlet-HQ/prod-evals-cookbook

    Production AI evaluations through 5 progressive stages — LangGraph evals, rubrics, failure mode analysis (Gauntlet AI Cookbook)

    Python

  3. hermes-gateway-selector hermes-gateway-selector Public

    Shell selector for choosing between local and remote Hermes Agent gateways

    Shell

  4. tanstack-spa-scaffold tanstack-spa-scaffold Public

    Greenfield TanStack SPA + Hono + Zod scaffold with agent quality gates

    TypeScript

  5. tool-repair-harness tool-repair-harness Public

    Hermes Agent plugin: deterministic tool-call repair harness for open models (DeepSeek, Qwen, GLM, Kimi)

  6. hermes-skill-math-via-code hermes-skill-math-via-code Public

    Hermes Agent skill: force all arithmetic through code execution

    2