> tom_butler

// full stack ai engineer · york, uk

AI you own,not AI you rent.

Private, local AI for businesses. Same results as the APIs for everyday work, no per-token bills, and your data never leaves the building.

See what it costs Talk it through
tom@local: ~

What I build

Three fixed-fee ways in. The audit is the cheapest way to find out whether the other two are worth doing.

AI Cost & Privacy Audit

What you are actually spending, what is actually leaving the building, and a straight recommendation. The place to start.

Local LLM Setups

Open models on your own hardware. Same results as the APIs for everyday work, with no per-token bill.

Private RAG Systems

Your own documents, searchable and answerable, with citations, and none of them sent to a third party.

See what it costs Read a case study

Why local AI

Most businesses rent their AI. Every API call is a meter running, and every prompt sends your data to a server you don't control. There is another way.

Renting it

per token
A meter that runs every time anyone uses it
their server
Your prompts and your documents, on someone else’s hardware
their terms
A processor to assess, and a retention policy to trust

Owning it

£0
Per-token cost, however heavily your team uses it
0 bytes
Sent to third parties. Nothing leaves the building
60-80%
Lower TOTAL cost at scale, once your own hardware and power are counted

Proof, not promises

The clearest way to judge whether I can build your system is to read how I built one.

case_study

Answering security questionnaires without leaking the answers

A private RAG agent that drafts supplier security questionnaires, grounds every answer in your own policy, and flags the ones a human needs to look at.

Read the case study

recently

  • Contributing to Odysseus: open source local AIAI2026
  • Sanctuary: fully offline on-device AI app (prototype)iOS2026
  • Starting Run It Local: a weekly plain-English AI newsletterWriting2026
  • Built RAG pipeline with LangChain + PineconeAI2025
now

Setting up private, local AI systems for businesses, and building Sanctuary, an on-device AI app for neurodiverse users heading to the App Store. Currently deep in Ollama, RAG pipelines, and open source local AI (I contribute to Odysseus). The theme of everything right now: AI you own, not AI you rent.

OllamaRAGLocal AIOn-Device
$ cat about.md

Ever since I watched The Matrix as a kid, I've been obsessed with building things on the web. This site is the sci-fi playground I always dreamed of, a space to experiment with AI, cyberpunk aesthetics, and creative code.

More about me GitHubTimeTravel
run_it_local

Run It Local

Keeping up with AI, and how to make it your own. One email a week, plain English, no hype. Written from the workshop floor, not the commentary box.

Subscribe one email a week · no hype