skillsync
Discover hidden talent in your codebase
375 followers
Discover hidden talent in your codebase
375 followers
Skillsync analyzes your codebase to find domain experts, discover unique skills, and capture successful patterns. No surveys needed—just timely insights that help you scale what works across your team.
Free Options
Launch Team / Built With
Render: Get up to $100K in startup creditsThe fastest way to launch and scale your startup in the cloud
Promoted
👋 hey product hunt!
open source powers the internet — but the contributors behind it are still underrated.
most platforms reduce effort to green squares and PR counts.
they miss the deeper signals and unique expertise of the contributors.
we built skillsync to change that.
it reads your real work — pull requests, reviews, comments — and builds a living skill graph.
not just what you did, but how you think, collaborate, and contribute.
with skillsync, you can:
discover hidden talent in your codebase
find the right experts for the right problems
scale unique skills with repeatable playbooks
even build your own agents on top of real team intelligence
we believe engineering is more than just shipping fast —
it’s about judgment, care, empathy, and craft.
skillsync helps make that visible.
try it on your favorite open source repo — or connect your own.
excited to hear what you think :)
https://sklsync.com
nars
co-founder, skillsync
This is a really powerful idea! I’m really curious about the AI-driven insights part - could this tool help even in early-stage projects where some contributors may still be learning or adapting?
@evgenii_zaitsev1 You bring up an interesting point because the number of lines of code each developer would be 'accountable' for is increasing exponentially even if the team size isn't. Things can quickly spiral out of control. Especially if a few team members are still learning, they could be brought up to speed pretty quick if you can create a playbook out of the desired skill levels!
I like this idea, @nars. How do you compare engineers based on the code they contributed? I.e. What differentiates a bad, ok, and great engineer with just that data?
Hey @rrhoover! We don't want to classify developers as a good / bad at all. This is why we don't have any proficiency / score that we assign people. skillsync is about finding possibilities in each contributor. And the first step towards that is knowing or empathizing with their work. We all have our natural inclinations. Some of us naturally enjoy 'firefighting' or pushing hot fixes and some of us enjoy deep research or creative work. What if we all deeply understood each others strengths and played to them? what if we could empathize at scale? That's what skillsync is all about.
skillsync looks at how people contribute differently. In fact we think skills are getting much more transferrable with ai, giving us more freedom to do what we like :)
This really hits cos SkillSync feels like the recognition open source contributors have deserved for years lol - What I like is that it’s not just always about what you build but also it’s how you think & this finally captures that!
Congrats to the team!!
@cranqnow I swearr! It's like there's so much good work out there to learn from yet it's barely used :)
@cranqnow If not now, will open source developers ever get recognized? Especially with the increasing usage of AI for writing code. To be honest, OSS deserves a lot of credit for how good AI is :)
Cool idea & Slick site — devs are going to love this! Curious how you're indexing all the repos — possible to share a quick overview of the data pipeline? Sounds like there's some fun stuff going on under the hood
@bernard_eugine
Okay so, instead of just snapshotting whole repos, we went way more granular—down to each PR and commit. Why? Because that’s where the real story lives. Early genius often gets refactored out later, but we catch it in the moment.
The real cheat code? PR review comments. That’s where devs drop gold—how they think, debate, make tradeoffs. You can’t see that in the final code, but we index it all.
Our system basically runs a timeline scrubber: walking through main branch history in order, picking up each PR like an episode in the show. It’s not just static code—it’s code with context, intent, and vibes.
Curious about the 'playbooks' feature. Could this help maintainers create better contributor guidelines based on what actually works vs what we think works?
@koushik_gampala Absolutely! The cool thing about playbooks is that the whole team does not have to rediscover the 'earned insight'. Today we can carry NASA's 10 principles for safety critical code as a system prompt with us to all our projects. So in this case the 'cost' of learning the contributor guidelines is approaching zero!
Hey Narayana, great idea and nice presentation! Do you think this product could be used to analyze "lack of skill" too to help upskill a team?
Hey@shashank_holla thank you! This one's tricky because we don't want to decide who's good or bad based on their skills. skillsync works best for when you want to scale the usage of an 'earned insight' across a large team :)