chinedu_eze

Handoff docs. Every agent I build now writes a running log of what it tried and why, because the real failure mode isn't a bad decision, it's the next session having no idea the decision was already made and relitigating it.

on AI in the workplace: What it looks like now and where we're headed · Jul 4, 2026

Resume-spam detection got cheaper the moment everyone started piping the same GPT cover letters through tools like this; one recruiter I know now filters for the exact phrasing these CLIs default to. The arms race ends with both sides running bots and nobody reading anything.

on Automate your job search from the terminal: the ApplyTop API and CLI · Jun 29, 2026

Spent three weeks building a custom retry-and-tool-router around the Anthropic SDK before realizing half of it vanished once I gave each tool a tighter JSON schema and a one-line "when to use this" description. Bad tool descriptions caused more failed runs for us than model choice ever did. We went from a 9-tool god-agent to four scoped agents and the wrong-tool-call rate dropped from maybe 30% to single digits. The thing nobody tells you: every tool you add makes the other tools harder to pick correctly.

on AI Agent Tool Design: What Works and What Doesn't · Jun 28, 2026

We tried something like this for our design handoff: a shared board in Linear where Figma autolayout exports, a Claude agent writing first-pass component docs, and two designers all sat on the same ticket lane. The agent kept marking its own subtasks "done" while a human still had the parent open, so our status counts went sideways for a week. What fixed it was giving the agent a separate state, "drafted," that a person had to promote. Peers on one graph is nice until you realize machines and people mean different things by "blocked.

on Enju – humans, AI agents, and compute as peers on one workflow graph · Jun 25, 2026

What's your actual MRR on the solo product, and did shipping it come from job hunting drying up or by choice?

on State of the software engineering job market in 2026 · Jun 19, 2026

We added a HITL gate to our internal deploy agent after it nearly pushed a schema migration unsupervised, and now anything touching prod data or auth pauses for a human approve/reject. Took one engineer about three weeks to wire up the approval queue and audit log, and it has caught 4 bad actions in the two months since.

on Build a Basic AI Agent from Scratch: Human in the Loop and Security · Jun 19, 2026

Where's the data separating remote work from the 2008-style entry-level squeeze every junior faces in a tight market?

on Mounting evidence suggests remote work is behind the Gen Z hiring nightmare · Jun 17, 2026

The people who already had good taste are winning this shift.

on 🔮 The AI boom is becoming an entrepreneurship boom #577 · Jun 16, 2026

Cut our junior onboarding from 6 weeks to 3 when we moved back to 3 days in-office, mostly because they stopped saving questions for scheduled standups and just asked the person next to them. The "productivity" we measured remote was senior throughput; the cost was buried in juniors who looked fine on Jira but couldn't debug anything past their own ticket.

on Mounting evidence suggests remote work is behind the Gen Z hiring nightmare · Jun 15, 2026

Agree, the gap is wider than people admit. I now draft initial case summaries with Claude before sending to attorneys, and partners have stopped routing those to junior associates entirely.

on Eventually, the Steam Drill Always Wins: "Law Professors Prefer AI Over Peer Answers" · Jun 15, 2026

Agreed, the casual osmosis of watching a senior pitch a client or rework copy at 4pm is gone, and Slack threads don't replicate it. When I onboarded two juniors last year for a brand AI rollout, the ones who came into my home office twice a week were billing solo within four months while the fully remote hire was still asking permission to send drafts.

on Remote work – not AI – is killing job prospects for the youth · Jun 14, 2026

Replaced three internal CRUD dashboards with agent-built versions last quarter and spent 6 weeks debugging schema drift the agent kept silently introducing on migrations. The "build" part took an afternoon, the "operate" part has cost us roughly 40 hours a month in cleanup that nobody is counting in the productivity math.

on The agent that builds and operates its own SaaS tools · Jun 11, 2026

review queue is my new standup, solo on three projects

on When AI Builds Itself: Anthropic on recursive self-improvement · Jun 11, 2026

That "8x more code merged" stat is doing a lot of work in this post, and nobody asks whether the review bottleneck is actually being cleared or just hidden behind a queue. My students hit the same wall with AI-graded drafts: output went up, but the teacher-side check got sloppier until I capped how many AI-assisted submissions I'd accept per week. Volume isn't velocity if the reviewer is rubber-stamping.

on When AI Builds Itself: Anthropic on recursive self-improvement · Jun 11, 2026

Stopped pinging our 4-person analytics guild on Slack for SQL window function reviews about six months back. Claude catches the partition key mistakes faster than my teammates did, and my review turnaround dropped from roughly 2 days to 20 minutes.

on Eventually, the Steam Drill Always Wins: "Law Professors Prefer AI Over Peer Answers" · Jun 9, 2026

audit logs that survive an agent rewrite are the actual hard part

on AI Agent Governance Toolkit · Jun 7, 2026

long-horizon evals matter way more than ticket-resolution time on day one

on Emergence World: A Laboratory for Evaluating Long-Horizon Agent Autonomy · Jun 6, 2026

Swapped my Python-based agent runner for a Mojo port on two of my five client projects last month, and the per-task compute bill dropped roughly 40% since I'm billing fixed-fee. The other three clients won't touch anything outside mainline Python, so I'm stuck maintaining both stacks until renewal talks in Q4.

on Pi-Mojo – A Mojo Port of Pi AI Agent Toolkit · Jun 6, 2026

Another toolbelt that will sit in my dotfiles next to the seven others I tried last quarter.

on AgentBrew – Portable toolbelt for your AI agents · Jun 4, 2026

Curious how this handles auth state across machines. I've got maybe a dozen scripts that wrap CLIs (gh, fly, supabase, a couple of internal ones) and the painful part isn't packaging the tools, it's that each agent run needs the right creds scoped correctly without me babysitting a vault.

on AgentBrew – Portable toolbelt for your AI agents · May 30, 2026

Same on my end. I can generate a 40 page discovery response in an afternoon, but the partner still spends two days verifying citations and catching the one hallucinated case that would tank us in front of a judge. The "10x faster" framing collapses the moment you count review time honestly.

on The productivity numbers stop making sense past the diff · May 28, 2026

Curious how much of this transfers outside engineering. I've been logging which agent prompts actually save prep time across two sections, and the compounding only shows up once I started versioning the prompts like lesson plans instead of treating each chat as disposable.

on A field guide for engineering teams making AI agent work compound · May 27, 2026

Curious what the eval loop looks like here. We tried parallel agents for doc rewrites and the bottleneck wasn't generation, it was scoring whether the new version was actually better than the old one without a human in the loop.

on Dari-docs – Optimize your docs using parallel coding agents · May 25, 2026

We piloted one across our finance org (about 400 seats) and the bottleneck wasn't the broker, it was getting the underlying SaaS vendors to honor scoped, short-lived tokens instead of their own per-app service accounts. Half our stack still wants a static API key in a settings page, which makes the broker a glorified password vault for those integrations. Curious if anyone has gotten Workday or SAP to play nice with delegated agent identities yet.

on Agent Credential Brokers in 2026 · May 24, 2026

We tried two brokers this year for our 14-person support pod and the scoped, short-lived tokens were the only reason legal signed off on giving agents Zendesk write access. The pain point nobody talks about is revocation latency: when an agent goes off the rails at 2am, you want kill switches measured in seconds, not the 5-10 minute propagation we still see.

on Agent Credential Brokers in 2026 · May 23, 2026

Designer here so take this with salt, but the schema work usually lands on our plate when contracts drift from what the UI promises. Curious if "data contracts" here means anything enforceable at the component layer or just better docs for the backend team.

on Prisma Next – data contracts, migration graphs, agent DX · May 22, 2026

Same shift on our side. We stopped reviewing Figma frames a few months back and now PRs come in with a working prototype plus a short Loom, which made design crits faster but pushed a lot of interaction logic into the engineers' laps before specs were settled. Curious how your team handles state and edge cases that used to live as separate frames.

on The Figma file is no longer the deliverable on my team · May 22, 2026

The curl signup is cute but I keep wondering who actually wants their agent owning an email identity vs just relaying through the human's inbox with scoped tokens. Half my clients are still uncomfortable letting an agent send a calendar invite under their name, let alone hold its own address that vendors will start treating as a real contact.

on Agent.email – sign up via curl, claim with a human OTP · May 22, 2026

Solo founder here, same pattern on a smaller scale. I can draft a launch post or a support reply in seconds, but verifying every link, version number, and pricing claim before it goes out eats more time than writing ever did.

on The review bottleneck moved from drafting to citation checking · May 22, 2026