Hacker News new | ask | show | jobs
Show HN: What leaves your company when an employee quits (skillpasspro.com)
2 points by kevinbaur 172 days ago
When someone leaves, companies think they lose a person. In reality, they lose context, decisions, shortcuts, and half-documented knowledge that no one notices until things break.

I built SkillPass after watching the same pattern repeat: “handover docs” that look fine but don’t answer the questions successors actually have.

SkillPass runs a single, guided session with the departing employee and turns their implicit, role-specific knowledge into a structured handover report, without meetings, training, or setup.

Free first handover, GDPR-compliant, no tracking, and nothing is used to train AI.

2 comments

Hey, I am Kevin the founder of SkillPass.

When I changed roles a few months ago I noticed something interesting. My successor onboarded in a fraction of the time it took me, simply because I was still around to explain everything. But during that offboarding I also realised how hard it actually is to pass on knowledge. I kept thinking, “Where do I even start?” and I was constantly afraid of forgetting important details. That is exactly what I wanted SkillPass to fix. A tool that gives people a clear thread to follow and asks the right questions so they do not have to figure everything out on their own.

Happy to answer any question! Kevin

I really like this and I can also see it being useful for one-off "state of play" of the environment, just thinking if the key resources from different teams all completed one of these how useful it would be for the rest of the company as well.

I had a question around the AI you're using, are you sending this off to OpenAI or doing this all locally?

Hi and thanks for the feedback, this is exactly the kind of “state of play” snapshot we had in mind initially. We also think there’s a lot of potential if key people from different teams each create one of these and make it available across the company.

Regarding the AI: at the moment SkillPass Pro is still using OpenAI, mainly because we’re in very early development and want to iterate quickly. Longer term, the plan is to move toward self-hosting and reduce external dependencies.

According to OpenAI’s API policies, data sent via the API is not used to train their models. The data is only retained for a short period and solely for service operation and abuse monitoring purposes, after which it is deleted. This was an important consideration for us when choosing this setup for the early stage.

Thanks again, Kevin

Will look at giving this a shot in the new year.

Thanks for confirming the AI provider, it would be good if they were listed somewhere on the site (but that's probably a personal preference). I'm a massive fan of local models, for your use case they'd be more than powerful enough to run on consumer hardware if you have some lying around. If you ever have any questions, feel free to drop me a DM, happy to chat about it.

This is an example of a WIP I've got in place that uses some spare local hardware available instead of using an external API - https://trulyanonymous.blazingbanana.com/survey/d14b4e2c-f44...

Thank you so much for your thoughtful feedback! I really appreciate it. The AI provider is actually already mentioned in our privacy policy, but I completely understand that it might not be easy to find. We're considering making that information more prominent on the site.

Thanks again for your willingness to chat and for sharing your local setup. It's always fascinating to see how different people tackle this! If you decide to give it a shot in the new year, I would truly value any feedback you'll have after you've had some time to explore it.

Merry Christmas, Kevin

Merry Christmas to you too!