•
Create one unpublished Dust agent on your own notes, test in Preview, and keep company Slack disconnected on night one.
Dust is a team workspace for specialized agents on company knowledge (docs, Slack, connectors). Home: dust.tt. First agent: Create your first agent. Academy: dust.tt/academy.
After work this shines for internal Q&A and drafting from your files—not for driving a local repo.
After-work Python learners who also keep notes in Notion/Drive/Slack and want a named helper for the team. If you only want a personal coding agent, use Amp or Cursor instead.
Overview: Top 10 AI agent tools for after-work Python.
Sign in at dust.tt. Prefer a personal or sandbox workspace. Do not dump employer secrets into a trial you do not control.
Home: Create. Start from scratch (or a small template). Sidekick in the builder helps draft instructions—treat it as a copilot for the agent spec, then edit.
Connect a folder of your public notes or a personal Notion/Drive you own. Least privilege. Tools: start with search over that source only.
Ask three questions you know the answers to. If it invents policy, tighten instructions and the data set.
Unpublished = only you. Do not share to the whole Slack workspace on night one.
Agent @study-brief: given a library name, return official docs URL, one-paragraph what-it-is, and three after-work steps. Ground it on a NOTES.md you upload. Call it with @name in Dust.
A named teammate over your notes beats a generic chat with no memory of the project.
When you want the same discipline in code, continue with Intro to Python Programming.