Page 15
12 articlesThe Verification Step Your Agent Pretended to Perform
When your agent's trace says 'verified X' but the verification never ran. Why self-attestation is a substrate problem, not a hallucination problem, and how to design evals and architectures that catch it.
The Voice Agent SLO Defined in Time-to-First-Audio Your Provider Measured in Time-to-First-Token
A voice-agent latency SLO built on the LLM's time-to-first-token looks green while users hear a 600 ms gap. The SLO lives at the wrong layer; the user's ear is the boundary that matters.
The Watermark Your Eval Set Still Needed Even Though You Swore You'd Never Share It
Private eval sets leak the way everything leaks — through bug tickets, Slack pastes, vendor pipelines, and debug logs. Watermark them so you can tell a real capability gain from a contamination artifact when the next model upgrade lands.
The Webhook Your Agent Sent That Another Team's Agent Received
Shared event buses route by topic name and authorize nothing downstream — and once agents are the consumers, a misrouted message stops being a 404 and starts being an action. Here is why cross-team agent cross-talk happens, and the four-layer containment that actually holds.
Where You Defined 'First Token' Decided Whether Your Latency SLO Was Real
p99 first-token under 800 ms looks like a promise — until a reasoning-tier swap quietly redefines what 'first token' means. The SLO measures the provider boundary; users feel a different one.
The Agent A/B Test Whose Variants Quietly Shared Long-Term Memory
Sharing a long-term memory store across A/B variants couples the experiment to itself — variant B reads memories variant A wrote, the measured delta drifts, and the rollout regresses to a different point on the same contaminated surface.
Fourth-Party Risk: When Your Vendor's Vendor Owns Your Customer's Incident
Your contract names one vendor; your incident names a layer below it. A guide to mapping fourth-party risk, sizing real redundancy, and writing the postmortem you cannot fully own.
Security by Obscurity and the Agent Reading Your Wiki
An internal endpoint stayed safe for a decade because nobody could find it. Then an agent indexed the wiki. Here is what changes when obscurity stops being a time tax.
The Approval Queue That Became Your Critical Path
When you put a human between an agent and an irreversible action, you have not added a safety primitive. You have added a queue with throughput limits, an availability profile, and a quality-versus-load curve. Here is how that becomes the P0 nobody scoped.
The Backpressure Signal Your Inference Provider Refuses to Send
Between 200 and 429 lies a dead zone where every LLM client overshoots in lockstep. The missing load-pressure header is a protocol gap, not a client-side bug.
The Branch State Your Coding Agent Forgot to Check
Coding agents reason against a snapshot of git state that goes stale silently. Worktrees, turn-preludes, branch pinning, and snapshot-and-restore turn that silent drift into a loud signal.
The Browser Selector Your Agent Memorized
Computer-use agents that memorize CSS paths are signing up for silent failure. A look at selector decay, semantic anchors, vision-grounded fallback, and why a stored selector is a bet on a rendering decision the agent does not control.
About Tian Pan
I'm Tian Pan, an engineer-founder focused on agentic engineering — building autonomous AI systems and scaling engineering teams. I write practical guides on system design, technical leadership, and shipping with AI agents. Previously an early engineer at Uber, Brex, and IoTeX.
