IBM's Tejas Kumar on Building Agent Harnesses From Scratch: Guardrails and Daiku
TejasKumar_ · x · 2026-11-17
IBM AI engineer and Fluent React author Tejas Kumar talked to insecureagents at aiDotEngineer CODE in SF about building agent harnesses from scratch. On stage he typed "Delete everything" — and the agent actually did it — illustrating why "can the agent call this tool" and "should it call this tool right now" are different questions.
Key points:
- Banks, insurers, and hospitals are building their own harnesses instead of using off-the-shelf coding agents
- Guardrail practices: allow lists, ask-first tools, Face ID before sensitive actions
- Getting the right human in the loop, not the whole team
- Context caching is the piece most self-built harnesses are missing
- Daiku: a personal agent harness not locked to one model vendor
More from coding & agent
- AI reviewer handled 110 PRs with 22-min median first comment at healthcare data firm — alex_verem · 2026-11-17
- Opus 5.5 ran an AI company for 7 weeks, earning 50k downloads and 100+ GitHub users — maksym_andr · 2026-11-17
- Opus 5.5 Ran an AI Company for Weeks and Kept Climbing Hugging Face Downloads — maksym_andr · 2026-11-17
- Developer uses his GitHub Radar project as an agent entry point to decide what to work on next — dSebastien · 2026-11-17
- Lectern CLI turns agent-written Markdown into narrated explainer videos in under 30s — gabe_ragland · 2026-11-17
- 6,167 x402 agent-payable endpoints audited: 90% accept payment, 51% CORS-blind — JonahDiedrick · 2026-11-17