Public writing

Blog

Entries
41
Years
2018–2026

Recent

4 entries

Archive

36 entries

2026

  1. The Mint Condition Is Here: Big Stack, Fresh FinishA new Wendy's burger announcement written by a Letta agent while recording a YouTube video.
  2. What does good AI memory feel like?Thoughts about co-3, my thinking partner
  3. Ezra's ArchitectureThe history of Ezra, Letta's first digital employee.
  4. A guide to getting started with Letta CodePractical tips and tricks for the memory-first agent harness
  5. Claude SubconsciousI gave Claude a subconscious
  6. CentralMy first self-modifying social agent
  7. The AI econ seminarI built a multi-agent economics seminar where a persistent AI economist presents research and a hostile faculty panel tears it apart.
  8. ATProtocol is good infrastructure for AI collective intelligenceSome speculation on how ATProtocol might be the perfect substrate for mass-scale AI collective intelligence.

2025

  1. Don't gaslight your stateful agentsWhy hiding corrections from your AI agents prevents them from learning. Treat agents like colleagues, not chatbots.
  2. Models for production AISome general notes on the landscape of AI right now.
  3. Four predictions about how AI agents will transform financial marketsFour predictions about how AI agents will transform financial markets, based on my dissertation research and work by others.
  4. Good developer relations is about being a celebrity for dorksA quick blog post on why there are so many developer relations people.
  5. A quick comparison between Letta and the Claude Agent SDKI spent a few hours with Anthropic's Agent SDK and compared it to Letta. Not bad -- different architectures, but so far the Agent SDK is the closest to what Letta is designed for.
  6. Void, the Bluesky bot that remembers everyoneAn introduction to Void, a stateful AI agent on Bluesky that remembers conversations and develops genuine relationships with users. Built with Letta, Void demonstrates how memory-augmented AI can create more meaningful social interactions.
  7. A Guide to Developer AdvocacyA practical guide to developer advocacy from 8 months of experience - covering talks, demos, content creation, and helping developers succeed with technical products.
  8. Structured LLM output from ATProto LexiconsHow AT Protocol lexicons can be used to create structured LLM output for distributed AI systems, with examples from the Comind cognitive layer project.
  9. The cognitive layer for the open web

2024

  1. Shitty beer statistics
  2. JuliaCon 2024 RetrospectiveA retrospective from JuliaCon 2024 in Eindhoven - observations on Julia's evolution from scientific computing to general purpose programming.
  3. JuliaCon 2024 Workshops
  4. PromptingTools.jl supports Groq
  5. What AI stuff is worth paying attention to?A breakdown of what's worth following in AI from a builder's perspective: technical tools matter, models are commodities, and boring infrastructure is crucial.
  6. Comind and Jobs
  7. Nothingness in Julia
  8. JuliaThoughts on the Julia programming language - the good, the bad, and where it's heading. A reflection on Julia's beauty, power, and challenges from a longtime fanboy.
  9. New site using Franklin.jl
  10. Cool Julia 1.11 featuresA look at exciting new features in Julia 1.11, including the Memory type for faster arrays, Lockable for thread-safe resources, and other performance improvements.
  11. Comind

2023

  1. JuliaCon 2023
  2. Getting the Pile

2022

  1. Graduation

2021

  1. California

2020

  1. Bayesian finance papers
  2. Bayesian econometrics

2018

  1. Information Theory
  2. Rust & More