Topics
- Claude
Anthropic's conversational AI model family used for coding and other tasks.
- AI-assisted development velocity
Measuring and scaling software delivery output with AI coding tools.
- Coding agent companies
Startups building AI coding agents and their competitive dynamics.
- Astra
OpenAI's frontier code model currently in use by engineering teams.
- Claude Code
Anthropic's coding agent that writes and edits code within Claude conversations.
- Jev
TypeSafe's general-purpose classifier model optimized for speed and cost over traditional LLMs.
- AI safety governance
Industry debate over pacing frontier AI development and third-party evaluators to reduce risks.
- Code readability
Concern that frontier models produce hyperefficient but unreadable code using meta-programming and obscure primitives.
- Inference optimization
Techniques and costs for matching model quality, speed, and caching performance.
- T3 Code
Theo's coding product measuring model traffic and usage.
Get the next one by email
Coding with AI, read daily so you do not have to. The experiments, the receipts and the arguments from engineers shipping real software. Not a changelog.