Boris Cherny's recent tweet featuring TLA+ in agentic coding has brought renewed attention to the 30-year-old formal modeling toolkit. The tweet, which used Opus 5.5 to model parts of the Claude Agent SDK in TLA+ and Lean, garnered significant online engagement, prompting many to inquire about TLA+.
TLA+ provides a compact language to describe possible system behaviors and the properties these behaviors should satisfy. While TLA+ checks a model of the software and explores finite instances, it does not fully verify an implementation directly. It serves as a starting point for verification efforts.
The discussion extends beyond TLA+ itself to how it can integrate with modern proof systems and AI agents. Tools like Verus allow specification, proof, and Rust implementation to coexist in the same language. AI can automate parts of this process; for example, an agentic pipeline converted over 16,000 TLA+ specification/property pairs into more than 3,000 machine-checked Verus proofs.
The key question is not just whether an agent can write TLA+, but what becomes possible when agents can consistently and reliably move between specifications, proofs, and real programs. Work is underway at Reasonable to train models that enable agents to perform these tasks, aiming for a loop of specified, implemented, and verified software.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's briefNew every morning, and the back catalogue is archived by date.
TLA+, a 30-year-old formal modeling toolkit, has recently gained internet attention after a viral tweet showcased its use in agentic coding. This renewed interest highlights TLA+'s role in describing system behaviors and properties, and its potential integration with modern proof systems and AI agents for software verification.