Tag

Customer Service Agent

14 posts

Make the Agent Get Sharper With Use, Not Dumber — Spinning Up the Data Flywheel

In labeling we found back-to-school promo questions spiked, but within them "how do I get the gift-with-purchase" was never answered well — so we fixed that KB specifically, and next round that class climbed. That's the data flywheel. In 5 minutes you'll see the first five posts are five spokes of one wheel; in 10 you'll spot why a flywheel spins in place (data piles up, never ships); in 20 you'll have a feedback loop ranked by frequency × cost that also forces you to split intents.

After Launch · 12 min read · EN · 中

A Pretty Accuracy Number Hid Dozens of Money-Moving Errors — How to Read the Eval to Ship

On a money-moving project I ran, the overall accuracy looked great; but pull the money-moving intents out on their own and the wrong-action rate was alarming — dozens of money-touching errors sat there the whole time, hidden by one blended number. In 5 minutes you'll see through "one accuracy figure to request launch"; in 10 you'll put a separate wrong-action gate on money-moving errors; in 20 you'll have a launch-decision flow: CI lower bound + per-scenario version cut + per-channel ramp.

After Launch · 13 min read · EN · 中

Your Dashboards Are Green While the Agent Quietly Gets Dumber — Post-Launch Silent Drift

Running a customer-service agent at a consumer-tech company, I learned one counterintuitive thing from watching the metrics: they're never a flat line. Every product launch, every back-to-school season, the question distribution shifts and a wave of new phrasings pours in — the agent quietly gets dumber, and not a pixel of it shows on the green CPU / QPS / latency dashboards. In 5 minutes you'll see through "dashboards green = healthy"; in 10 you'll have 6 leading signals that fire weeks before complaints; in 20 you'll turn the eval set from "frozen at launch day" into one that re-samples current traffic.

After Launch · 13 min read · EN · 中

Your Labels Are Your Ceiling — One "Swap Half a Size Up" Gets Three Answers From Two Agents

A customer's one line — "I want to swap half a size up" — hides four calls: exchange or return, intercept the shipment or not, refund the price difference or not, and against which price. Two skilled agents label the same 50 rows back-to-back and agree on only 35 — your 96% accuracy was measured with a ruler that's only 70% self-consistent. In 20 minutes you'll have a flow for measuring agreement first, then writing rules like "which price the difference is refunded against" into a rubric.

After Launch · 13 min read · EN · 中

50 Rows at 96%, Ship It? Size Labeling by the CI Lower Bound, Not the Pretty Number

"We labeled 50, 96% correct — ship it?" No — the statistical lower bound is only 86%. When I designed the labeling gate for a customer-service agent, the first rule I pinned was: read the confidence-interval lower bound, not the point estimate. In 5 minutes you'll see through the small-sample 96% mirage; in 10 you'll have a "true accuracy → rows to label" table that explodes near the threshold; in 20, a CI stopping rule — label more where it's needed, don't waste it where it's done, and call a stop when the answer is fix-the-agent.

After Launch · 12 min read · EN · 中

The 96% Turned Red the Moment We Split It by Channel — Two Fatal Failures of Eval Sampling

The vendor drew 200 rows at random and reported "96% accuracy." I split those 200 by channel — in the small public-domain channel, the price-difference cases were nearly all wrong, and the high-traffic private channel had diluted it into a pretty number. In 5 minutes you'll see through both failures of random sampling (missing the tail + a big channel drowning a small one); in 20 you'll have a stratified frame with channel drill-down and risk-layer over-sampling.

After Launch · 13 min read · EN · 中

Corpus Drives Codebook — Why Your Intent Taxonomy Is Stuck at 60% and How It Evolves from 36 to 48 | Agentic AI in Practice (X)

Customer-service Agent in production, 36 intents, unknown rate 40%, the business side asks 'can we just add an LLM fallback?' The real problem is not the classifier — it's the codebook itself. Five minutes in you can spot the wrong diagnosis ('unknown rate high = classifier weak'); ten minutes in you have the four-quadrant test that filters 80% of pseudo-missing-intent requests; twenty minutes in you have the corpus → codebook iteration loop that evolves a taxonomy from 36 to 48 stable intents.

Agentic AI in Practice · 14 min read · EN · 中

Don't Let AI Agents Call APIs Directly — A 5-Layer Tool-Calling Stack + 25-API Contract Checklist

The most common fake architecture in customer-service Agent projects this year: 'we let Agents call order / ticket / logistics APIs directly, 25 integrations done, full coverage' — then ask 'what happens when the OMS vendor changes?' answer 'rewrite,' 'how does QA do mock integration?' answer 'wait for the real interface,' 'compliance audit for write operations?' answer 'we'll add logging.' This is missing layers. Written for architects, founders, and project owners running enterprise Agentic projects: 5 min to spot the most expensive architecture mistake, 10 min to lock in the 5-layer responsibility split (Adapter / Service ABC / Tool / Workflow / Critic), 20 min to walk out with a 6-systems × 25-APIs integration matrix + 5 architecture decisions to drive this week.

Agentic AI in Practice · 17 min read · EN · 中

Intent Classification for Chatbots: Why Pure-Rule and Pure-LLM Both Fail (a 3-Tier Cascade)

Intent classification is the first node in any customer-service Agent — get it wrong and the next four architecture decisions are wasted. Pure-rule is brittle; pure-LLM blows the budget. The 3-tier fallback (rule → embedding → LLM) is the only engineering trade-off that stands up. Five minutes in you can spot the two fake architectures ('just use an LLM' / '100% rules'); ten minutes in you have starting thresholds for all three tiers; twenty minutes in you have the signals that say it's time to evolve from HybridClassifier to LLM Router.

Agentic AI in Practice · 16 min read · EN · 中

Pytest-Green Doesn't Mean Ship-Ready: How to Actually Test an AI Agent (Dual-Track)

The thing your customer-service Agent project gets most easily fooled by this year: 'pytest 400+ green, coverage 79%, CI gate passing.' Then the boss asks 'what's the faithfulness rate? Tone compliance? Prompt-injection block rate?' and nobody answers. The 'tests passed' bar for an AI system is not the 'tests passed' bar for traditional software. This piece is for architects, founders, and project owners shipping Agentic AI inside an enterprise: 5 min to see why pytest-green is misleading, 10 min to decide who owns which 4 of the 8+ test buckets, 20 min to walk out with a 7-quality-dimension threshold table + 3-cadence rhythm + 5 things to drive this week — bring it to your next architecture review.

Agentic AI in Practice · 16 min read · EN · 中

Containment Rate vs Resolution Rate: The Only Customer-Service AI Metric That Matters (How "98% CSAT" Gets Faked)

The CEO gets a weekly email from the vendor: CSAT 98%. I pulled the raw data — ~5% of customers rated 'satisfied,' a fraction of a percent rated 'unsatisfied,' 95% never responded. 'Silent = satisfied by default' is how that 98% got built. Five minutes to see through four flavors of fake-resolution claim; ten minutes to redraw your team's customer-service north star.

Agentic AI in Practice · 16 min read · EN · 中

Why a 70% Critic Beats a 95% Critic — A Fail-Closed Design Deep Dive | Agentic AI in Practice (III)

A Critic second-pass review is the only thing standing between an L2 customer-service Agent and a refund mistake. But the '95% automation rate' vendors keep showing you is almost always fail-open — Critic times out, the action passes through. Five minutes to see through three flavors of fake backstop; ten minutes to redraw your team's design.

Agentic AI in Practice · 22 min read · EN · 中
← All posts