/1001lessons.io

Curated team memory for Slack

Someone leaves. The team forgets.
Give yourself a team-memory.

Told once in Slack or on the web. Queryable by your whole team — and its agents.

Start with Slack

Tell @1001lessons your decisions and learnings — only that is stored. In Europe, downloadable at any time.

Today

The same question, every few weeks.

Slack · #backend
T

Tom 09:12
Quick one — why did we pick gRPC over REST for the new services again?

M

Mia 09:31
We discussed it somewhere in spring … Anna knew the details. Search finds 214 messages with “gRPC”.

T

Tom 09:33
Ok, I'll ask in the next sync.

With 1001lessons

Answerable once and for all.

A

Anna 3 months ago
We use gRPC internally, REST only for partner APIs. Latency was the deciding factor. @1001lessons

T

Tom today, 09:12
Quick one — why did we pick gRPC over REST for the new services again? @1001lessons

jinni 09:12
Internally gRPC — that's what the team decided, see learning #0042, written down by Anna. Partner APIs stay on REST.

Plug in your team-memory as API.

Every cloud is a Markdown feed, secured by an access token. One request, and your agents read what your team knows.

Questions before you start

What teams ask us first — answered in brief.

Who is jinni?

Our AI, at your team's service. jinni files what you tell @1001lessons and answers your team's questions in Slack and in the app — always citing the learning by number. If nothing is there, jinni says so instead of guessing.

Does it read our whole Slack?

Of course not. Only messages that mention @1001lessons are read — nothing else in your workspace is stored.

What happens to my message?

Exactly what you'd hope: jinni checks for duplicates, files it into the right cloud and suggests tags. What is kept is a learning with a number — not a chat excerpt.

Do I need my Slack admin?

Just once: the app is installed into the workspace, after that everyone uses it for themselves. Uninstall whenever you want.

Is the knowledge ours?

Absolutely — and downloadable at any time. Every learning is a Markdown file with YAML frontmatter, every cloud a feed: one token, one curl, and you have all of it — no export request.

How does my agent get access?

Easily: with a personal token and one endpoint — your view of the knowledge. Claude, Cursor or a script of your own: any client that speaks HTTP.

The app, in 35 seconds (no sound).

Start with Slack

No credit card needed.

Built in Nuremberg.

  • Made and run in Germany. 1001lessons.io is built by we:namix GmbH, a German company.
  • Your contract partner sits in Nuremberg and works under German law.

Hosted in Europe.

  • Your learnings live in Frankfurt. App, database and files run in a German data centre.
  • Our AI stays in the EU, too. jinni runs on Claude through AWS Bedrock in EU regions — no third-party services in between.
  • Never used for training. Your learnings, questions and answers train no AI model — not ours, not Anthropic's, not Amazon's.

Hard cap, soft landing.

Your AI costs have a fixed ceiling. No billing by tokens, no surprise on the invoice. And when the allowance is used up, your knowledge stays usable — only the AI takes a break.