My TOP 3 traits for recognizing the potential in engineers: - sense of ownership and responsibility, - drive and motivation for improving, - team player and making everyone around them better.

Sponsorship slots on the Engineering Leadership newsletter are now fully booked 4 months in advance. Just a couple of slots left are either in discussions or semi-available.

OpenAI is using a new interview format: a product + technical deep dive. Candidates tackle an AI-native problem and collaborate with interviewers to design the product and system in real time. No prework, just solving the problem together.

Want quality? Shorten feedback loops: - Better tests - Smaller PRs - Faster reviews - Clear ownership Quality is a workflow.

As an engineer, you can do the best work in the world, but if the expectations are not met correctly, all the effort is wasted. First build the RIGHT things, second build the things RIGHT.

The next era of leadership is less motivation, more environment design. You don’t “inspire” deep work. You remove friction. You set fewer goals. You kill recurring meetings. You make quality the default. Culture is just the shape of the calendar.

Execution is getting cheaper. Judgment is getting expensive. Anyone can ship code faster now. What will separate leaders is choosing the right problem, saying no to the wrong ones, and protecting the team’s attention long enough to finish something that actually matters.

The best engineering strategy is simple: Pick fewer things. Finish them. Ship them.

Saying "I don't know" is a sign of seniority for me.

Just asking for a friend :) Happy Friday and have a great weekend!

The best engineering leaders won’t “know more.” They’ll know what to ignore. AI will flood your org with options, dashboards, opinions, and half-baked ideas. Your job becomes subtraction: fewer priorities, fewer meetings, fewer “nice-to-haves.” Clarity will be the rarest skill.

I can tell a lot about engineer's seniority by the way they approach solving problems. Are they really taking care of the problem? Or do they just accept the first solution they come across and leave it for others to deal with?

The next era of leadership is less motivation, more environment design. You don’t “inspire” deep work. You remove friction. You set fewer goals. You kill recurring meetings. You make quality the default. Culture is just the shape of the calendar.

AI makes output cheap, impact becomes rarer. So, anyone can "ship more code", while business impact becomes that much more important.

Software engineering is becoming position-less. Titles are becoming less important. Impact is becoming everything.

Waiting for some pokemon AI tools to put it in the list :) Happy Friday and have a great weekend!

The ultimate leadership flex is making yourself unnecessary. Not by disappearing but by building a team that can decide, ship, and work well without you. If everything needs your approval, you’re not leading. You’re the bottleneck.

When AI makes output cheap, impact becomes rare. Anyone can ship more code. The real multiplier is the person who helps the whole team ship better code, focusing on the right problems, with fewer failures.

Advice for engineering leaders building AI-native engineering teams.

Being able to understand users, ask the right questions, and decide which problems are actually worth solving matters more than ever.

Being able to understand users, ask the right questions, and decide which problems are actually worth solving matters more than ever.

As a leader, when you trust your people and empower them, in many situations, you’ll be surprised by what kind of amazing things they’ll come up with.

As a leader, when you trust your people and empower them, in many situations, you’ll be surprised by what kind of amazing things they’ll come up with.

As a leader, when you trust your people and empower them, in many situations, you’ll be surprised by what kind of amazing things they’ll come up with.

What AI-native engineering teams do well is: Identifying where work slows down and applying AI to remove those bottlenecks.

What AI-native engineering teams do well is: Identifying where work slows down and applying AI to remove those bottlenecks.

What AI-native engineering teams do well is: Identifying where work slows down and applying AI to remove those bottlenecks.

An interesting thing that came up talking with @sulmanrc, Head of Engineering at OpenAI, is that they like to hire engineers in 2 extremes: 1. Really great generalists 2. Someone that is an expert in a really specific thing and think outside of the box

The best engineers ask better questions. Not because they’re smarter. But because they respect the cost of building the wrong thing.

Gotta catch (learn) 'em all! Except it's AI tools instead of gym badges :) Happy Friday and have a great weekend!

The teams that succeed in the AI era aren’t the ones using the most tools or moving the fastest at any cost. They are the ones who communicate well, understand each other’s pressures and priorities, and keep learning together.

As an engineering leader, if you’re the smartest person on the team, you’ve failed. Your job is to build a team that solves problems without you. Otherwise, you’re not “hands-on”, you’re the bottleneck.

Trust comes from visibility. Show the numbers. Show the risks. Show the trade-offs. People can handle bad news. They can’t handle surprises.

Use AI coding tools. But use them to HELP you, not to do the work FOR you. Opening a PR filled with AI-generated slop is the fastest way to lose credibility.

Use AI coding tools, but use them to help you, not to do the work instead of you. Just opening a PR filled with AI-generated slop is a fast way to lose credibility.

It happens 🤷‍♂️😄 Happy Friday and have a great weekend!

Great chatting with @thsottiaux from the OpenAI Codex team. Interesting deepdives on Codex team and AI-native engineering teams coming up on the Engineering Leadership newsletter.

No matter how great technically you are, weak relationships and low visibility will limit your influence and opportunities.

Your manager shouldn’t ask you for a status update. Instead, you proactively provide the update. I recommend using this template (thread):

Most people believe being reliable is something that comes as a side effect of doing good work. But they are wrong:

5 phrases that build trust fast: 1. Tell me more. 2. I might be wrong. 3. Help me see it your way. 4. Here’s what I can do today. 5. Thank you for pushing on this.

Question for you: Are you hiring more engineers in 2026, yes or no? If no, why?

How to become a pragmatic engineer: 1. Deliver iteratively 2. Ask better questions 3. Reduce scope without reducing value 4. Build proof-of-concepts and prototypes 5. Focus on good-enough solutions and avoid perfectionism

Doing great work is only half the work. Making sure everyone knows about it is the other half. You need to do both.

Doing great work is only half of the work, making sure that everyone knows about it is the other half. You need to do both.

Tech Lead is one of the most desirable roles in tech. The reason? A mix of technical and people skills, including: • Common sense and pragmatism • Good communication and leadership • Being a multiplier for others, making everyone around them better

The best engineers ask better questions. Not because they’re smarter. But because they respect the cost of building the wrong thing.

Meetings don’t create alignment. Decisions do. If nothing was decided, you wasted everyone's time.

Most engineers are making this mistake when using AI tools for coding. (You are probably making it too):

Typing speed was never the bottleneck in software engineering. The bottleneck was always thinking and picking the right thing to build. AI just made that painfully obvious.