Stripe built its internal Kai knowledge agent in one week with one engineer and Deep Agents
LangChain · x · 2026-08-04
What happened
LangChain says Stripe built Kai, its internal Knowledge AI Platform, on the Deep Agents harness in one week with just one engineer.
Why it mattered
Stripe had multiple teams already building agent layers on top of Ruby and Java systems, but production-grade agents needed more than connectivity:
- Reliability at scale
- Better performance
- Strong evaluation and regression handling
- Security and access control for non-engineers
How Stripe approached it
- Stripe’s AI platform team built an in-house harness rather than forcing everyone into developer tooling
- Kai became a company-wide productivity agent
- The stack is based on LangChain/LangGraph plus Deep Agents
- The article frames the effort as a way to let employees build agents without needing terminal-heavy workflows
More from coding & agent
- Stanford’s CS239A course on self-improving AI agents is now on YouTube — mariofilhoml · 2026-08-04
- Cline says open-weight models work better when the harness lets them verify more — cpaik · 2026-08-04
- WorkOS schedules Agent Night in San Francisco with context-graph talks and live demos — davidcrawshaw · 2026-08-04
- A curated reading list for DeltaNet, FlashKDA, vLLM serving and MoE — austinvhuang · 2026-08-04
- Developer builds browser FM synth editor with DeepSeek 4 Flash for $1.10 — keunwoochoi · 2026-08-04
- Yann LeCun says strong code-generation systems go beyond plain autoregressive LLMs — ylecun · 2026-08-04