Tom Walczak

Hi, I'm Tom

Building verifiable, truth-seeking AI agents

I'm building AI agents that verify their reasoning on complex, controversial questions — surfacing counter-arguments, stress-testing logic, and turning confusion into clarity. This will 100x intellectual work and make real-time fact-checking possible.

I'm based in London, UK.

What I'm Building

Open Debate

Open Source

An AI debate system for stress-testing arguments. AI can get very good at presenting, challenging, and evaluating reasoning. Bad arguments are built on weak foundations. When pressed, they fall apart.

  • • Recursive self-improvement for AI debaters
  • • Vanilla baseline configuration for surfacing model bias
  • • Human-in-the-loop coaching mode
  • • CLI-first, integrates with Claude Code for analysis
View on GitHub

Argument Analyzer

Open Source

Real-time argument analysis as you write. Two AI agents work in the background: Steelman surfaces the strongest version of your argument, Counter shows the objections your opponents will raise.

  • • Feedback appears as you write, no button to click
  • • Click feedback to jump to the relevant passage
  • • Chat assistant for improving specific sections
View on GitHub

AlexAI

Conversational AI embodying energy expert Alex Epstein's thinking. A production AI system that captures and communicates ideas from Fossil Future and The Moral Case for Fossil Fuels.

  • • 900+ indexed sound bites, timeless principles
  • • AlexAI Pro — Research-oriented tool with stricter guardrails
  • • Commentary Engine — AI-powered news analysis
Try AlexAI

Blumberg AI

Built an AI agent for Blumberg Capital that screens incoming pitch decks, scores deals against investment criteria, and syncs with the CRM — making the VC process faster and more founder-friendly.

Read the case study

Recent Writing

View all →

Head of AI at Alex Epstein | AI Consultant