The Superpowered Individual (Combination-of-Skills Career Moat)
Go deep in one craft, then use AI to add adjacent crafts until you become un-replaceable
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 90%
A career strategy for the AI era: pick one domain and go genuinely deep in it, then use AI to become good enough at the two adjacent domains so you can do the whole job yourself. The value comes from the rare combination, not any single skill. Use it when deciding how to invest your learning hours as a PM, engineer, designer, or any specialist worried about being automated.
Origin
Andreessen describes a 'Mexican standoff' between coder, PM, and designer where each now believes AI lets them do the other two roles, and argues they're all correct. He grounds it in Scott Adams' famous career advice (being a cartoonist who understood business made him spectacularly great at Dilbert) and Larry Summers' economist framing ('don't be fungible'). Lenny reframes it as a T-shape; Andreessen extends it to an 'E or F on its side' (deep in one, competent in several).
Core principles
- 01The additive effect of being good at two things is more than double; three things is more than triple
- 02Depth in at least one domain is non-negotiable — it's how you judge what the AI hands back
- 03Being merely one thing makes you a swappable cog; a rare combination makes you un-replaceable
- 04AI is the lever that lets a single person pick up the adjacent skills that used to require hiring others
How to run it
- 1
Go all the way down in one domain
Pick your primary craft (coding, design, product) and build genuine depth — for a coder that means understanding the stack down to how the AI and the microprocessor actually work. This depth is what lets you evaluate and correct AI output.
Pro tip Depth is what converts AI from a black box into a tool you can direct and debug.
Watch out If you only know how to prompt but can't read what the AI produces, you can't tell good output from mediocre output.
- 2
Use AI to add the two adjacent domains
Once deep in one role, deliberately use AI to become competent at the neighboring roles — a master coder learns to also do AI-driven design and product management, and vice versa.
- 3
Aim for the rare combination, not more of the same
Target the intersection that few people occupy (the 'auteur' who can write and direct, the 'unicorn' who can code and design). The combination is what makes you a super-relevant specialist.
Pro tip Ask: what pair of skills would make me one of the only people who can do this whole thing?
- 4
Stay un-fungible as tasks shift
Keep expanding scope so your value lives in orchestrating the whole product, not in any single task an AI can absorb. The goal is to be the person who directs the AI across all three domains.
Watch out Don't rest on the deep skill alone — a master coder who never learns to leverage AI will eventually hit a ceiling.
In the wild
Adams said he was only a pretty-good cartoonist and pretty-good at business, but the fact that he was a cartoonist who understood business made him spectacularly great at making Dilbert. Neither the world's best cartoonist nor the world's best businessperson could have produced it.
→ One of the most successful cartoons in history, created purely from a rare skill combination.
Plenty of writers can't direct and plenty of directors can't write, and both can be successful. But the superstars who move the field are the ones who can write and direct — the 'auteurs.'
→ The combination, not the single skill, produces the field-defining creative force.
Common mistakes
Staying a pure single-role specialist
If you're 'just a designer' or 'just a coder,' you can be swapped in or out. Larry Summers' warning is 'don't be fungible' — a lone skill makes you a replaceable cog.
Skipping depth and going straight to breadth
Trying to use AI across three domains without deep mastery of one leaves you unable to evaluate or fix what the AI gives you, so you never reach the 'superpowered' tier.
Is it for you?
Best for
Specialist ICs (PMs, engineers, designers) who fear their single role will be commoditized and want to compound their value in the AI era
Not ideal for
People early in a career who haven't yet built depth in any one craft — they should establish the deep leg first before spreading wide
From the transcript
“the additive effect of being good at two things is more than double. The additive effect of being good at three things is more than…”
“the fact that I was a cartoonist who understood business made me like spectacularly great at making Dilbert”
“what that means essentially is don't be replaceable”
From the episode
Marc Andreessen: The real AI boom hasn’t even started yet