proof of work used to be a great way to showcase your skills beyond a cv or resume.
but now? they’re all the same. the same generic, user research, same generic ICP documentation, even in devs, the same cloned github projects. whenever a copy paste option exists, 99% will do it.
99% of these docs are useless. literally useless. so then what changes in the 1%? for the next few days i will be covering proof of work. across all functions. i’ll also be talking about building with ai and building for cracking ai roles.
but before we do that
the reality of any business is that it’s a business. that means it’s trying to grow users or revenue or both. take an example of a seed stage founder.
this founder needs to grow revenue from $100k to $200k this quarter. they're interviewing candidates.
one shows a blog post about growth strategies.
another brings certificates from a growth bootcamp (yuck!).
the third walks in with a working prototype that reduced their checkout abandonment by 12% after testing with 15 actual customers.
the gap between what professionals build and what companies need is brutal. most portfolios are college projects with better typography. they demonstrate skills in isolation. growth modeling in a vacuum. ai capabilities without context. design thinking without users.
it’s proof of value. something that creates immediate value. not tomorrow. not after three onboarding weeks. monday morning.
this time let’s simplify it even more?
and say three chefs apply.
slides you a 20 page pdf. "my philosophy on optimal cupcake creation." complete with process diagrams. flavor matrices. temperature charts. you flip through it. academic citations about gluten development.
you know what you're thinking? this person has never worked a friday night rush. next.
second candidate. pulls out their le cordon bleu certificate. mentions they graduated top of their class. shows you photos of their final exam: a croquembouche that took 14 hours. beautiful. useless.
every candidate has certificates. some fancier than others. you'll chat because you're polite. but you're checking your phone under the table.
third candidate. no slides. no certificates. opens a box. inside: three variations of your bestselling red velvet cupcake. first one uses 20% less cocoa, same taste profile. that's 5L saved annually. second has a cream cheese swirl that takes 30 seconds to add but lets you charge $2 more. that's 2L/month revenue. third is a limited edition with gold leaf that creates urgency. tested it at a local market. sold out in two hours.
"i talked to 15 of your regular customers last week. they said your cupcakes are good but predictable. here's what they actually want." now we're talking.
who are you hiring? the first two showed competence on paper. the third showed up having already done the job. not proof of work. proof of value.
the third chef for your target role.
1/ pick 2 to 3 companies where you want to work. real companies with real problems. you’ve already done this if you’ve been following the series. you probbaly also know where to build proof of work (hint, it’s where your gaps lie.)
if you haven’t done this, do that before you pick the proof of work.
2/ what’s their growth equation. every business reduces to a math equation. revenue equals aov times number of orders. number of orders equals new customers plus returning customers. new customers equal sessions times conversion rate.
pick one lever. just one. conversion rate dropping? time to value too long? churn spiking after day 7? choose something measurable.
3/ talk to their actual users. not your assumptions about users. not personas from a blog post. real humans who pay real money. find 10 to 15 of them.
post on social: ""who uses [company x]? i'll buy coffee for 15 minutes of your time.""
linkedin, reddit, discord. wherever they are. go there. book 15-minute calls. ask about their experience. their frustrations. please restrict this to THE LEVER your solving and not about the company.
study the market they operate in. what are the constraints? what's changing? where are the opportunities everyone else missed?
not a figma file. not a strategy deck. a working product that moves that one lever you picked. something they can implement without translation.
use the skills you're trying to prove. building for an ai role? create an ai solution. growth role? build a growth experiment. make it functional. if it needs code, write code. if it needs automation, build automation.
show your prototype to 5 of the customers you interviewed. document their reactions. measure the impact. would they use this? would they pay more for this?
everything else is vanity metrics for your peers.
create a brief presentation. the problem. your solution. user feedback. projected impact on their metrics.
send directly to the hiring manager. not hr. not the application portal. find their email. (we’ll cover how to distribute proof of work later in the series)
this changes the entire conversation. you're not a candidate anymore. you're a someone who already solved their problem. the interview becomes a discussion about implementation, not evaluation.
i've seen people do interview rounds as a formality because their proof of work was so good. the interviews became formalities. ""as long as they don't completely fuck up, they're hired."" that's the bar we're aiming for.
not to test you. to learn from you. product wants to understand your user research. engineering wants to see your implementation. leadership wants to know how you thought about the problem.
this creates dialogue.
even leaders need this. you think directors pow doesn’t work? i've seen them get hired based on it. abhinav from IIT delhi, then IIM ahmedabad. two tier one credentials. still built specific proof of work to crack a director role at pocket FM. used the gx framework. got hired.
this approach takes 10x more effort than building a portfolio website. you'll talk to customers who don't want to talk. you'll build prototypes that don't work the first time.
but once you do this, you could probably reuse the approach for anyone in their industry. and honestly the question to be answered here isn’t if your portfolio is impressive.
does your proof of work already work?