AI agent engineering & essays
◆
The Book
of Life
of Life
❧
Building reliable systems,
and thinking beyond them
and thinking beyond them
On AI, building, philosophy, psychology,
football, chess and travel
football, chess and travel
Robat Das Orvi
I build software and write about AI agent engineering: production reliability, tool security, state, and spending controls. I co-created OneThread and AgencyHandy. This site also holds 154 essays on building, thinking and football.
About meLatest essays
- Replay Production Prompts to Test an LLM Model Router
- Loss Aversion in SaaS Pricing: Why Most Paywalls Backfire
- Accept an Intended API Schema Change Without Losing History
- API Pagination Cursor Expired Mid-Run in an Agent Tool
- AI Agent OAuth Consent Phishing: Blocking Rogue App Grants
Start here
AI agent engineering
Practical writing on what happens when agents use tools, make decisions, and run in production.
Explore the AI agents chapter- Why LLM Agents Fail Silently in Production (And How to Detect It)A silent agent failure doesn't crash. It reports success. Here's the uncomfortable economic reason most teams never find out.
- Least Privilege for AI Agents: Scoping Tools, Keys and FilesystemsLeast privilege for AI agents in practice: how to scope tools, credentials and filesystems, with the Replit and Amazon Q incidents as evidence.
- Durable Execution for AI Agents: Temporal vs a Plain State StoreDurable execution for AI agents: when Temporal earns its keep, when a plain state store is enough, and why resuming can be the dangerous part.
