June 19, 2026

AI for the Over 40 – Week 27: Building Your Own AI Career Development Partner

What if performance management evidence was collected by you throughout the year instead of about you during annual review season?

Last month, I sat down to complete my self-evaluation. I’ve been through some version of this exercise for more than two decades across multiple companies. As I stared at the screen, I realized something uncomfortable: I was going through the motions.

The competency frameworks didn’t reflect how I actually spend my time. I was scrambling to recall accomplishments from eight months earlier, and I found myself writing for an audience rather than honestly reflecting on my development.

Then I went looking for my goal worksheet from earlier in the year. It had disappeared during a system migration. And the most revealing part? I felt relieved.

That reaction told me everything I needed to know.

The question I couldn’t stop asking

Across multiple companies, I’d seen the same pattern repeat: annual goals that get set and forgotten, competency frameworks disconnected from real work, year-end scrambles to reconstruct accomplishments, and ratings that feel more performative than developmental.

The intentions behind these systems are usually good. Organizations want fairness, growth, retention, and meaningful feedback. But somewhere between intention and execution, the process breaks down.

If even people who care deeply about development experience performance management as administrative overhead, the problem isn’t motivation. It’s the system’s design.

Managers are overloaded. Documentation becomes reactive. Development conversations happen too infrequently to create real change. Over time, we normalize the dysfunction because everyone assumes this is simply how performance management has to work.

The paradigm inversion

Traditional performance management collects evidence about you. Managers observe your work, form impressions, and eventually synthesize those impressions into a rating. You discover where you stand when someone else tells you.

What if we inverted that model?

What if evidence was collected by you throughout the year? What if you tracked your own trajectory weekly instead of reconstructing it annually? What if the final rating wasn’t a surprise, but a confirmation of what you already knew?

That changes the evaluator’s role completely. Instead of acting as the sole judge of your performance, managers become calibrators who ensure consistency across teams and peers. That’s a fundamentally different relationship.

The system I built starts with a simple principle: every year begins at “Developing.” Not because someone lacks capability, but because performance must be demonstrated through documented evidence over time.

That shifts the psychology completely.

In the traditional model, people assume they already own a rating and spend the year defending it. In the new model, you build a case for the level you want to achieve, one piece of evidence at a time.

If you go a year without documenting meaningful evidence, you can’t realistically defend anything beyond “Developing.” That’s not punishment. It’s accountability. And it eliminates surprises.

What AI makes possible

A human manager cannot realistically provide weekly developmental attention to every person on their team. The math simply doesn’t work. So most organizations default to quarterly or annual conversations that are too infrequent to meaningfully shape behavior.

AI changes that equation.

I built a system that lets people collect evidence weekly from email, calendars, meeting transcripts, completed work, and personal reflections. An AI assistant helps analyze that evidence against development dimensions, identify patterns, highlight gaps, and track trajectory over time.

The AI isn’t replacing human leadership. It’s enabling a level of continuity and attention humans alone can’t consistently provide at scale.

And the constraints of building it taught me something important: the hardest problems weren’t technical. They were behavioral.

How do you make reflection feel useful instead of burdensome? How do you encourage honest self-assessment without turning it into performative reporting? Those are design problems, not coding problems. Increasingly, I think those are the problems that matter most in enterprise AI.

The accountability shift

This experience reinforced something I’ve argued throughout this series: you cannot delegate your career development any more than you can delegate your AI transformation.

Both require ownership.

The evidence collection isn’t primarily for managers. It’s for you. The trajectory tracking isn’t for HR reporting. It’s for self-awareness. Conversations with your advisor become better because you’ve already done the work of reflection before entering the room.

That changes everything.

Your Week 27 challenge: audit your own development

This week, take an honest look at how you currently approach your own growth.

1. What evidence actually exists?

If your annual review happened tomorrow, what documented evidence could you immediately point to? Not what you remember doing — what evidence actually exists?

2. Who is doing the synthesis?

Are you actively tracking your own growth and trajectory, or are you expecting someone else to notice your development and explain it back to you later?

3. What would weekly attention change?

Imagine spending 10 minutes each week documenting meaningful work, lessons learned, and growth signals. What patterns might emerge that annual reflection completely misses?

4. Where are you outsourcing responsibility?

Career development is deeply personal work. Where have you been waiting for someone else to own something that ultimately belongs to you?

The bottom line

I sat down to complete a self-evaluation and felt relieved when I couldn’t find my goals. That reaction revealed how disconnected the system had become from genuine development.

So instead of improving the form, I started rethinking the model itself.

What emerged was a fundamentally different approach: evidence collected by the individual, continuous reflection rather than annual reflection, AI supporting awareness and synthesis, and managers serving as calibrators rather than surprise evaluators.

The deeper lesson extends beyond performance management.

You cannot delegate your own growth. Not in your career. Not in AI adoption. Not in personal transformation.

The people who benefit most from AI won’t be the ones waiting for systems to change around them. They’ll be the ones building awareness, capability, and accountability before anyone tells them to.

This post is part of my “AI Over 40” series. It first appeared on LinkedIn: AI for the Over 40 [Week 27]: Building What Didn’t Exist: An AI Career Development Partner

Read more AI and Copilot blogs.

Trending Posts