Someone builds a reporting automation, a follow-up flow or a small internal app. It works for them. But the next person may not know it exists, where the code lives or what they need to use it.
AffinityAI is the shared memory for how your company works with AI. Your preferred AI tool is still where you do the work. AffinityAI keeps the useful approach, its source, who knows it and what a teammate needs to get started.
Our starting principle is simple: don’t document, save what works. Let the connected assistant structure the useful details, and keep people in control of what they share. Start small: one automation, one colleague, one useful head start.
Good workflows keep evolving.
Someone creates a useful workflow. A colleague adapts it. The original improves. Those relationships should stay understandable without anyone learning version-control terminology. Ask what changed, compare it with your own version and choose what to carry forward.
Our pilot now brings that memory into ChatGPT cards, with Slack discovery and a small Updates view in the library. The ambition is broader: as agents and tools change, the team’s accumulated know-how should remain useful.
Useful work should outlast
a tool or a team change.
Teams change providers. Colleagues change roles or leave. The automations they built should still be understandable to the people who come next. Keeping source references, setup, ownership and history together makes the next move easier to plan.
That shared knowledge should be yours to take with you, including if you leave AffinityAI. Our test workspace now exports saved workflow records with their source references and optional history, so the next colleague or AI assistant has a starting point outside the library.
Continuity also needs company access to the underlying source and services. AffinityAI helps keep that knowledge visible; transferring accounts, preserving code and adapting platform-specific parts still need to be arranged.
Hi, I’m Mikael.
Most people call me Miksu.
I’m a software entrepreneur and the founder of AffinityAI. I’ve built and led companies, worked with growing teams and learned plenty from things that didn’t go to plan.
I’m also building AffinityAI with AI involved throughout the work—from product thinking and design to development and research. That makes this a very practical question for me: how can a useful way of working become something other people can use, too?
AffinityAI began with collaborative scenario planning. This new direction focuses on the automations and practical know-how people create while doing their jobs, and helping someone else find and reuse them across tools.
I’m always happy
to compare notes.
If your team is finding useful ways to work with AI, I’d like to hear what’s helping and what keeps getting lost.
— Miksu