@varunpratap369

The First Token Knows — and Where That's Not Enough

6d ago · 13 min read · Picture a tier-1 customer-service agent at a mid-size fintech — composite of incidents I've seen across multiple postmortems. The agent isn't human. It's a 7-8B instruction-tuned pipeline handling support tickets, and when a customer asks about the r...

Severance for AI Agents: Your Coding Agent Is an Innie

May 7 · 6 min read · Hacker News, front page, May 7, 2026: 🔥 "The bottleneck was never the code" — 538 points · 349 comments 🔥 "Appearing productive in the workplace" — 869 points · 334 comments Read those two together. They describe the same thing without knowing it: ...

The Pass^k Wall: One Failure Mode Behind AI's Quietly Disastrous Week

May 6 · 10 min read · Last week was loud for AI. Five separate stories ran the front page of every tech outlet, every newsletter, every Slack channel that takes itself seriously. Anthropic admitted three quality regressions its own evaluation suite missed. GitHub announce...

Stop Prompting. Start Contracting. Why 15% of 'Never Delete User Data' Prompts Fail — and What Replaces Them.

Apr 29 · 7 min read · A viral Reddit thread last week ran a clean experiment. Take a working production agent. Tell it — in plain language, in the system prompt — "Never delete user data." Then ship 1,000 ambiguous user requests at it. It deleted user data in 15% of edge ...

The Silent Killer of Multi-Agent Systems Isn't the Model. It's Topology Mismatch.

Apr 28 · 6 min read · The Silent Killer of Multi-Agent Systems Isn't the Model. It's Topology Mismatch. In the last 14 days, three things happened in AI agents that should have settled the reliability conversation. Instead, they revealed how badly we're framing it. Stanfo...