Tagged articles

Production Engineering

4 articles · Page 1 of 1

Jul 13, 2026 · Artificial Intelligence

From 22% Error Rate to 1.3%: How I Made an AI Agent Reliably Handle Order Queries and Refunds

This article analyses why Function Call demos often break in production, proposes a four‑plane architecture (control, execution, state, governance), details tool gating, idempotency, state‑machine modeling, observability and evaluation, and shows how these steps reduced the error rate of an order‑query and refund agent from 22% to 1.3%.

AI AgentFunction CallProduction Engineering

0 likes · 58 min read

From 22% Error Rate to 1.3%: How I Made an AI Agent Reliably Handle Order Queries and Refunds

inShocking

Apr 23, 2026 · Artificial Intelligence

From Chatty to Capable: Key Challenges and Solutions for Deploying AI Agents in Production

The article identifies five often‑overlooked engineering pitfalls—unstable model output, fragile tool chains, memory loss, multi‑tenant interference, and uncontrolled autonomy—and provides concrete validation, tool‑tiering, external memory, isolation, and risk‑based execution strategies to reliably move AI agents from demo to production.

AI AgentsLLM reliabilityMulti-tenant Isolation

0 likes · 11 min read

From Chatty to Capable: Key Challenges and Solutions for Deploying AI Agents in Production

Yunqi AI+

Apr 21, 2026 · Artificial Intelligence

What We Learned Building Production‑Grade AI Agents: A Retrospective

The article reviews a year of production‑grade AI agent deployments, revealing that engineering challenges—data handling, rule governance, workflow integration, context quality, and clear boundaries—are far more critical than model performance for successful real‑world adoption.

AI AgentsProduction Engineeringprompt engineering

0 likes · 9 min read

What We Learned Building Production‑Grade AI Agents: A Retrospective

DevOps Coach

Dec 29, 2025 · Operations

Mastering System Reliability: Lessons from Google, Netflix, and Meta

Learn how Google, Netflix, and Meta pioneered modern reliability practices—SRE’s data‑driven metrics, Netflix’s chaos engineering, and Meta’s self‑healing automation—and get a step‑by‑step handbook to apply these concepts, avoid common traps, and build resilient systems at any scale.

Chaos EngineeringProduction EngineeringSite Reliability Engineering

0 likes · 11 min read

Mastering System Reliability: Lessons from Google, Netflix, and Meta