My answers to career advice people have asked me.
Should I go deep or broad in my career?
Early in your career, sample actively, i.e. go broad. Commit to depth when process (not just outcomes) energizes you. Read Range (David Epstein, 2019).
How do I explore when I’m stuck in a role?
Say yes to stuff you don’t like. You’ll do what you like anyway. For breadth, you need the opposite. You’ll learn:
- New things you enjoy. (Do you enjoy the same things you enjoyed 10 years ago?)
- How to enjoy stuff you don’t like. (A useful meta-skill.)
- How to get out of stuff you don’t like. (Another useful meta-skill.)
How do I make boring work more fun?
- Do the minimum on boring tasks. Spend more time on fun adjacent tasks.
- Make boring work more fun. (I once wrote an RFP so dry no human could read it. So I turned it into a 430-character poem via the first letter of every page.)
- Don’t assume what’s boring now will be boring later. (Testing bored me until Simon Tully explained that it’s detective work!)
What skills should I learn?
- Ignore your passions. Pick what’s rare and valuable.
- In winner-take-all markets (e.g. sports, sales, research), double-down on the most important skill. Otherwise (e.g. consulting, entrepreneurship), stack related skills to build rare + valuable combinations.
- In rapidly changing fields (e.g. technology, digital marketing), learn how to learn fast. In stable fields (e.g. law, medicine), gain knowledge.
- In your early career, bet on emerging technical skills. In late career: invest in slower-decay judgement & leadership skills.
Will AI use atrophy my skills?
Yes, but that’s OK. You can’t do long division as well as we did before calculators. But it’s less important today. Instead, you analyze several KPIs as ratios.
- AI will definitely atrophy some skills - but many of them will be used less. Observe which ones you’ll need and why.
- AI will also necessitate new skills. Over-use AI to discover what these are. Under-use of AI is a bigger risk than skill atrophy.
What projects should I work on?
Choose projects where the process of getting good at them solves a problem you care about for people whose judgment you respect.
| Short Runway | Long Runway | |
|---|---|---|
| Low Capital | Skills-first. Optimize for learning. Pay « Best mentors. | Explore widely. Many small bets, diversify |
| High Capital | Leverage & monetize, preserving optionality. | Moonshots: high risk + upside. Startups, books, research. |
- For fast industries (tech, AI, media): short, intense explorations since skills depreciate faster.
- Slow-clockspeed industries (law, medicine, infrastructure): specialize deeply since skills last longer.
- Regulated industries (pharma, finance): Build credibility with gatekeepers. Invest in compliance, quality, and trust-building.
What if my projects aren’t deployed?
- Unshipped work is often harder work - more ambitious - which risks failure. Recognize its value over an easy, safe bet that’s deployed.
- Hard work is often niche work. Competitors often hire for such niche experience.
- Try to get it deployed. It teaches you organizational navigation and resilience.
- Sometimes, projects aren’t meant to be deployed. They’re for business to experiment and learn.
- Share project failures in your CV. “We built this, and it didn’t go anywhere” builds your credibility.