From Senior Engineer to Staff, with Angelic Assistance

Or: How to get promoted on purpose, with robots

A friend and reader has challenged me to write about just how one rises beyond the level of senior software engineer to the more rarified air of staff/principal/distinguished/fellow/whatever-your-org-calls-it. How one goes from helping to build a product to helping to build an organization. How one gets the opportunity to truly make an impact via technology, rather than management.

I’ll happily write this post, rather than pointing them at many good ones out there, because right now is possibly the best time ever to attempt to rise beyond senior software developer. Folk with the technical skills, business effectiveness, and personal agency required of staff/principal/distinguished engineers are needed more than ever as organizations attempt to integrate the strange new angels of agentic AI. Simultaneously, the tools available to you have changed, enabling you to ship with more impact, learn more with more breadth and depth, and communicate with more reach than ever before. For extra bonus nachos, your competition probably hasn’t noticed yet. Let’s do this.

The Three Pillars

Promotion frameworks for staff+ engineering positions usually boil down to two things: expertise and visibility. Be extremely good at what you do, and be known to be extremely good. I’ve given that basic advice to any number of mentees. It is true as far as it goes, but is incomplete.

The missing third pillar is intentionality — the meta-skill of building a promotion plan deliberately rather than accumulating skills and accomplishment and hoping someone notices. For anything you care about, the worst advice in the world is “Just be yourself”. That most definitely applies to your career.

Here’s what this series of posts will cover.

Expertise Development: How do you get good at something, fast enough to matter? How do you absorb complex bodies of technological and business knowledge continually. How do you pick which thing to become expert at? What does “expert” even mean at senior-plus levels where everyone is already pretty good? How can AI tools be leveraged for the sort of aggressive continual learning that is required of higher engineering ranks.

Visibility Development: How do you get known? What’s the difference between visibility that helps and visibility that annoys? How do you build a reputation inside a company where you might be remote, new, or organizationally distant from decision-makers?

Doing It Deliberately: How do you put it together? How do you make a plan, stress-test it, and adapt when reality doesn’t cooperate? How do you surface hidden assumptions, analyze the impact of unexpected changes, and re-orient toward success. It is here where the agentic AI tools really show their strengths as a genuine force multiplier for your planning and execution loop.

Why now, and what do robots have to do with it?

The last eighteen months have seen an explosion of agentic coding tools — Claude, Cursor, Copilot, Codex, Cline, Conductor and that’s just the C’s. The engineers who matter are using these to write code faster, and sometimes better. That’s fine. It’s the obvious application, and a valuable one.

But code velocity isn’t the bottleneck at senior-plus levels. The bottleneck is scope — understanding systems you didn’t build, navigating organizations you don’t know, producing impact that travels beyond yourself and your team.

Agentic tools change this equation:

For expertise: You can now analyze codebases at a depth that used to require months of immersion. You can use specialized analysis agents that surface patterns, history, and relationships that would otherwise stay buried. The moat around “I’ve been here five years and know where the bodies are buried” just got a lot shallower. This applies to the structure of your business as well, making it easier than ever to learn the levers that actually drive value in your organization and speak in terms that the higher levels of your business understand. Almost as an afterthought, agentic learning tools have made expertise in new technologies nearly too cheap to meter.

For visibility: Using AI tools, you can produce more sheer evidence of your expertise — proof-of-concepts, architecture documentation, technology proposals, blog posts, tech talks — at higher quality and higher frequency. The constraint shifts from “do I have to I write this up?” to “should I write this up?” More touch points. More chances to get seen. More chances for one higher-up to send an e-mail to another entitled “Is this the guy we need for <X>?” AI is impacting marketing nearly as much as it has impacted software development. No reason not to use it to market yourself.

For intentionality: You can use AI as a planning partner, stress-testing your strategy and identifying blind spots. This is not at easy as it sounds. It requires a detailed understanding just what you’re trying to accomplish and what you are capable of. Anyone can use AI to answer questions. The real lever here is using it to question your answers. With AI, you can build the sort of plans that let you react with suppleness and force as conditions change, exploiting opportunities and mitigating risks faster than you would have thought possible. You can approach your career with analyses, not vibes.

The people you’re competing with for that Staff slot? Most of them are using AI to autocomplete their React calls. A few are using it to amplify their scope. Be in the second group.

The Dogfooding Disclosure

I’m not writing this from the summit. I’m writing this from the climb.

I’ve bounced around between principal/distinguished/fellow for about twenty years now, depending on where I’ve been working. When my previous company got acquired seven months ago, I dropped from distinguished to principal. Totally reasonable of them, and not something I’ve got ego about, but I certainly don’t intend to stay at principal, particularly at sixty. As such, I find myself working on exactly this problem, yet again. I’ve built the plan. I’m executing the plan. This post, the first in a four-part series, is me documenting what I’m learning in something close to real-time.

Why admit it?

First, honesty. I’m not going to pretend I’ve cracked the code when I’m still running the experiment. I like you all, but not enough to bother to lie to you.

Second, writing from inside the problem is a more useful frame for understanding task. Promotion advice from someone who got promoted twenty years ago and now dispenses wisdom from on high is... okay. I can certainly point you to some good and well-written items in that genre. What I can offer instead is the live playbook — what I’m trying, what’s working, what’s not, with updates as I learn. I’ll be sharing what I can, down to the level of individual prompts, although I will need to keep some hole cards secret.

Third, the tools themselves are new enough that nobody has twenty years of experience using them for career development. We’re all figuring this out. I just happen to be writing it down.

One Weird Trick To Ace Your Promotion … NOT!

There are no guarantees in any of this. Promotions depend on factors outside your control: budget, timing, politics, your manager’s relationship with their manager and the manager above that, whether the company is growing or contracting and which parts of the company are growing or contracting. All you can do is maximize the part you control and be smart about the part you can’t.

At the career levels we’re talking about, it’s not about gaming the system. It’s about understanding the system well enough to do genuinely good work and make sure that work is legible to the people making decisions.

Coming Up

Next post: Expertise Development. How do you become the expert in something, deliberately? How do you pick the right something? How do agentic tools change your learning curve? When you can learn with more breadth and depth and alacrity than ever before, how do you manage your expertise portfolio?

I’ll be drawing on some war stories, but mostly I’ll be focused on the tactical: what to do this month, this quarter to build the kind of expertise that gets recognized at senior-plus levels.

If you’re planning your own senior-to-staff push in 2026, I’d love to hear what you’re working on. The comments are open, and I read and respond.

For me, 2025 was a train wreck crashing into a dumpster fire and then plunging into a plague pit. I achieved my victory condition, mere survival, but it’s time for something more. Join me, as we write our names on our organizations in letters of fire.

---

This post was constructed as always with the able assistance of Claude Opus 4.5. I shudder to imagine what Claude’s promotion packet would look like.