What Idempotent Actually Means: Why Retries Are Safe (and When They Aren't)
Originally published on Loop & Retry — field notes on building LLM agents that survive production. Search for "what does idempotent mean" and yo…
Latest News
⚑ Report a ProblemTech news from the best sources
All topics
AI
Gear
News
Tech
agents
ai
api
architecture
automation
beginners
career
database
devchallenge
devops
javascript
llm
machinelearning
mcp
opensource
performance
productivity
programming
python
react
security
showdev
testing
tutorial
typescript
webdev
Your Backup Doesn't Exist Until You've Restored It
Everyone has backups. Almost nobody tests them. And an untested backup is not a backup — it's a hope, wearing a backup's uniform. I learned this the…
Production Hardening an AI Video Pipeline: Retries, Fallbacks, and Crash Guards
The video generation platform I work on orchestrates a long chain of AI calls — a video generation provider for clip rendering, Claude for scene pla…
"The write was acknowledged" means less than you think
I spent years operating databases for payment platforms, where a lost write isn't a bug, it's a regulatory conversation. That environment teaches yo…
Your agent's p99 is a different animal
Originally published on Loop & Retry — field notes on building LLM agents that survive production. The demo felt instant. The agent answered in…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
Streaming tool calls without losing your mind
Originally published on Loop & Retry — field notes on building LLM agents that survive production. Streaming exists so the user isn't staring at…
Retry patterns: when you should give up (and why most code doesn't)
Originally published on Loop & Retry — field notes on building LLM agents that survive production. The retry-budget post answered the HOW: a sha…
Preventing Duplicate Password-Reset Notifications (Under SMS Timeout and Retry Pressure)
Treat an SMS timeout as an unknown outcome, not a failed send: accept each password-reset event once, persist its expiry and idempotency key before…
A Free Model Endpoint Replied Twice, Then Went Silent. The Fix Was a C++ Replay Envelope, Not Retries
Late on a Tuesday, a C++ tooling team noticed their warning classifier was duplicating classifications. The batch runner sent forty compiler diagnos…
One bad step, N bad steps: how agent failures cascade
Originally published on Loop & Retry — field notes on building LLM agents that survive production. Here's the failure mode that surprises people…
Building Trust Through Transparency: Why Mifugo Added a Public Status Page
When you're building software for farmers, you're building software for people whose livelihood depends on it. A feed log missed at 6 AM isn't just…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
The Silent Outage: Monitoring What You Can't See
The worst kind of outage is one nobody notices. Your metrics are green. Your dashboards are fine. Your users are quietly getting a broken experience…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
Webhook Retries Aren't Optional: The 6-Hour Silent Failure That Changed How We Build Them
We found out our webhook delivery was broken the same way most teams do: a customer told us. Not an alert, not a dashboard going red — an email aski…
How Do You Measure AI Agent Reliability?
Your agent passed the eval, so you shipped. The next day a user sends almost the same input and it fails. Nothing changed. You just learned that "it…
Test OmniRoute Fallbacks for Semantic Consistency, Not Just Availability
diegosouzapw/OmniRoute is an MIT-licensed AI gateway that advertises one endpoint across many providers and models, with quota-aware fallback and to…
Test Saga Compensation When Payment Succeeds but Inventory Times Out
An order reserves inventory, charges payment, then times out waiting for reservation confirmation. Cancellation begins. A late inventory success arr…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
If 30% of Coding Tasks May Be Broken, Your Leaderboard Needs an Uncertainty Budget
OpenAI published an audit of SWE-Bench Pro on July 8, 2026 and estimated that roughly 30% of its tasks are broken. The reported issues make a famili…
AI memory가 자기 출력을 다시 입력으로 세지 않게 하는 법
AI memory가 자기 출력을 다시 입력으로 세지 않게 하는 법 AI memory 시스템은 대화에서 기억을 만들고, 그 기억을 다시 요약해 더 높은 수준의 결론을 만들 수 있습니다. 여기서 원본에 가까운 입력과 시스템이 만든 파생 결과를 구분하지 않으면, 파생 결…
Automatic Error Recovery in AI Agent Networks
In a single-agent system, failure is simple: the agent errors, you retry. In multi-agent systems, failure is a graph problem. The Cascade Failure Pr…
Canary Agentic Autofix With Failure Classes and Reliability Gates
GitHub announced agentic autofix for code scanning alerts in public preview on July 10, 2026. Primary source: GitHub Changelog, July 10, 2026 . The…
Merge Concurrent Agent Patches by Base Commit and Hunk Ownership
Two coding agents start from commit A. One finishes quickly and merges as B. The second finishes later with a patch that was valid against A. Applyi…
Reject Agent Tasks Before Queue Age Breaks the SLO
A queue can accept work long after it has lost the ability to finish that work on time. “Message accepted” is not availability when the task's deadl…
Long-Running AI Agents Need a Lease, a Heartbeat, and a Drain Test
NVIDIA's June 2026 GTC Taipei program includes a session on long-running AI agents that plan and operate over hours or days. Once a task outlives on…
Treat Per-Task Model Switching as a Concurrency Protocol
Changing the model for a running AI task is not a settings update. It is a distributed operation: read current task -> prepare credentials/config…
Service Level Objectives for Complex Microservices
Why SLOs Break in Microservices A SLO that works for a monolith often collapses when you distribute the same logic across 30 services. The math of a…