Principal Engineer · 16+ years
Architecture, evaluations, observability, failure handling, cost and latency: the parts a demo never has to answer for.
Or go deeper: the live cohort ↗For directors, heads of function and founders: eight live evening sessions with Ehsan Gazar and Anoosh Zeighami, ending with a one-page 2027 AI plan for your organisation, built on a real workflow and ready for your leadership team.
Starts Wed, 4 Nov · 7:00pm London · live on Zoom
$1,495, or $1,045 with code EARLYBIRD until 23 Oct.
Enroll on Maven ↗Free, live, 30 minutes, for organisational leaders. AI may have just settled a 90-year-old maths problem, yet the same systems still fail simple tasks. A manager's framework for capability, intelligence and reliability, so you know when your teams should trust an AI output and when they should not.
Free, live, 30 minutes, for organisational leaders. Take one knowledge workflow apart and redesign it around AI: which old constraints still shape it, what AI changes, what must stay human, and how to run a first bounded experiment.
Also on Maven, waitlist open: Production-Ready Systems with LLMs and Agents ↗ · From Senior to Staff: Master the Architecture Skills That Get You Promoted ↗
29+ engineers and colleagues, sharing what changed after we worked together.
“It has been a pleasure working with Ehsan. He is a dedicated, self-motivated and knowledgeable software engineer. He always embraces new challenges at work. He has a positive attitude and desire to produce quality work that serves him well. He would be a valuable team member of any software team.”
“Ehsan gave me really clear and practical advice on my resume and career direction. His feedback helped me focus on what actually matters to recruiters and improve how I present my skills. Super helpful and easy to talk to !!”
“I've been working with Ehsan in Netbarg. He has always proved to be a results-oriented teammate who has great technical skills. I would like to recommend Ehsan as a key co-worker to whom to trust any critical work.”
An open-source terminal for running coding agents. See what each one changed, read the diff, then stage and commit without leaving it. No account, no telemetry.
See Marlin →One TypeScript problem, written from memory against hidden tests, on a schedule that brings it back just before you would have forgotten it. No account needed to try one.
I use AI every day, I enjoy it, and I am not going back. What it quietly took is the part of me that produces code from an empty file: the best practices I used to apply without thinking, and the problem solving underneath them. Why the existing practice sites did not fit, and what I built instead.
Your agent gave a bad answer and you cannot say why. Tracing tells you what happened on one request. Evals tell you whether quality moved across many. They answer different questions, and if you only build one this quarter, build the one that makes the other actionable.
Clayton Christensen showed why good engineering teams build irrelevant products. The paradox of listening to your best customers, and how to stop ignoring the toy technologies that will eventually replace you.
Nearly every agent framework decided that one agent means one container. Cloudflare says there are not enough computers on earth for that to work, and has flipped it around: the agent lives in a tiny sandbox with a filesystem that stays behind, and a container is only borrowed when Linux is truly needed. Here is the argument in plain words, why Cloudflare made it, and where I would use it.
Agent graphs are state machines with a new name on them. The engineering question is not what a node is, it is which routing decisions you take away from the model and which you leave to it. The rule I apply: only draw edges for decisions you can make better in advance than the model can at runtime.
Some questions cannot be answered by adding one more join, because the number of joins depends on the data rather than on the query. That is the moment your data has become a graph. Here is how to tell, what a graph actually buys you, and the two costs nobody mentions until you are already paying them.
What other engineers have been reading in the last seven days.
A short, simple review of Leopold Aschenbrenner essay. He says AI will reach human-level smarts around 2027. Here is what he gets right, what he gets wrong, and what it means for you.
Lefthook replaces Husky and lint-staged with a single Go binary, parallel hook execution, and built-in file filtering. Here is why I switched and kept going.
Replaced ESLint with oxlint in a React TypeScript project and cut lint time from 12 seconds to under one. What I traded away and whether it was worth it.
A few months ago I was reviewing an incident postmortem where a payment service took down the entire checkout flow for 40 minutes. The root cause wasn't a...
Most engineers write HLDs and ADRs in markdown. Here is why I switched to HTML and React specs, and how living documentation changed my design reviews.
Staff engineers are judged on whether a problem was worth solving, not just whether they solved it. I do not find those problems by thinking hard. I collect them, let them pile up, and wait for the shape.
The deep dives I'd point a senior engineer at first, grouped by topic.
Worked SQLZoo solutions, joins, aggregates, GROUP BY vs HAVING, and subqueries — the queries interviewers actually ask.
9 deep dives · Start reading →
Indexes, query optimisation, sharding, replication and normalization — how to make a database fast and keep it fast.
8 deep dives · Start reading →Why oxlint replaced ESLint, pnpm vs npm, Lefthook vs Husky, and the Rust/Go shift in the frontend toolchain.
6 deep dives · Start reading →
Arrays, linked lists, stacks, queues, trees, tries, graphs, heaps and hash tables — implemented and explained in TypeScript.
12 deep dives · Start reading →
Insertion, selection, merge and quick sort, plus binary and substring search — the classic algorithms, step by step.
6 deep dives · Start reading →
What a design system really is (per Atlassian), how Dropbox and Netflix build theirs, and how to roll out an internal UI component library.
6 deep dives · Start reading →
Horizontal vs vertical scaling, capacity estimation, resilience patterns and the architecture principles behind systems that grow.
6 deep dives · Start reading →
Head-to-head breakdowns — oxlint vs ESLint, REST vs GraphQL, SQL vs NoSQL, Svelte/Preact vs React — with the trade-offs that decide it.
12 deep dives · Start reading →
