"We trace everything" is an intention, not a guarantee: one new code path is enough to open a hole nobody sees. What holds the rule is a test that freezes the inventory of call sites — plus two traps that cost me time.
"The agent hallucinates" is a bad diagnosis: it sends you looking at the model. On an agent extracting data from documents, I watched the guarantee move — from model quality to three engineering constraints. One of them surprised me.
I kept editing the system prompt, kept re-running, and nothing moved. The culprit wasn't a cache: the graph is built once at startup. Three lifetimes coexist inside an agent, and knowing which one applies changes the feedback loop.
Once you delegate to agents, you always have two or three branches in flight. A plain `git checkout` breaks everything: stash, Docker stack restart, ports fighting each other. Here's the layout that unblocked me — and what it actually costs.
An entire access-control design rested on a line of code that doesn't exist. The code wasn't wrong — the ground was. Here's why a fabricated line number is an agent's most effective lie, and the guardrail that catches it.
42 sessions, 913 messages, 155 commits. The report doesn't show what I thought I was doing — it shows what I actually do: hand over broad missions, let them run, and correct mid-flight instead of specifying upfront. Here's the loop, and what it costs.
They say it takes 14 steps to become an agentic AI engineer. Spoiler: real transformations fit into 5 architectural patterns. Here's how I learned it's no longer about talking to the model, but building a system that decides.
Going from one agent to two isn't about writing twice the code. It's about adding a third piece that never answers: the orchestrator that detects intent and routes to the right specialist.
Freelance developer based in Réunion Island, +2h from Paris, fully remote. Spoiler: the time difference changed nothing. What mattered was the backend architecture I laid down for a team that had to scale — DDD, microservices decomposition, guidelines that hold.
An LLM predicts text; an agent runs tasks. In between sits the harness. I set out to understand it, and realized my mental pattern already described the Claude Agent SDK.
The dev job is shifting from typing to steering: orchestrating agents, writing the spec, guaranteeing that what ships is what was expected. What changes when you stop coding to produce, and start coding to validate.
An AI agent ships «average» code by default: 300-line files, any everywhere, logic in the JSX. Here are the 10 rules I wrote for VoiceJournal, and why they hold.
Going from Claude that codes to Claude that disciplines itself. SessionStart to bootstrap the repo, PreToolUse to block sketchy commands, Stop to notify you when a task ends. Four concrete hooks running silently in my repos.
A GitHub issue from the phone, a Claude Code session auto-triggered in a cloud container, a PR popping up in Slack, a merge from the GitHub app. My indie workflow when I'm away from the laptop — and what works vs. what doesn't.
Brainstorming → spec → detailed plan → sub-agent execution with review. The workflow I've been using for a few months that finally makes AI usable on multi-day projects, not just snippets.
A Rust CLI proxy that sits between Claude Code and your shell commands. It compresses verbose output (git status, ls, jest...) before it ever reaches the context window. One command to install, instant payoff.
A markdown file at the root of your repo describing how we code here. Read by Claude (the agent) before every session, and by humans at first glance. It's become my most important engineering contract.