All micro-internships
Micro-internship brief ≈15h across ~3 days — get real users before submittingbuilder

Ship a Telegram bot people can actually message

Ship something with real users’ front door: a Telegram bot, live on a public handle, that does one genuinely useful thing — daily reminders, a quiz, mess-menu lookup, expense logging. AI drafts the code; you understand it, fix it, and ship it.

AI Application DevelopmentConversational AI

Posted by The AI Internships

The work brief

  1. 01Create a bot with @BotFather (free) and pick one useful job for it.
  2. 02Build it with AI-assisted code (Python/Node) — host it free (Render, Railway, PythonAnywhere, or long-polling from Replit).
  3. 03Get at least 2 real people to use it and keep their (anonymised) feedback.
  4. 04Push code to a public repo and record a demo video of a real conversation.

What you’ll produce

4 deliverables

Submission standard

Submit the bot’s public @handle, the repo, a demo video link, and your build story. We may message the bot — it should respond. Include your AI workflow: tools, best prompts, and what you changed from the AI’s output. Your repo must show the bot built over more than one commit, and the @handle must respond when we message it.

  • Your bot’s public @handle and what it does

    Short answerRequired

  • Public GitHub repo

    Public GitHub repo — commit history should show the bot evolving (handlers, hosting fixes) across more than one commit, not a single paste. A one-commit dump fails the authenticity gate.

    Public linkRequired

  • Public demo video of a real conversation with the bot

    Public linkRequired

  • Build story: hardest bug, what your 2 real users said, what you’d add next

    Written responseRequired

You’ll complete these inside your private workspace.

How your work is evaluated

The passing benchmark is 70/100.

Actually shipped

29%

A real bot exists, responds, and does its stated job (per demo).

Code quality

14%

The repo contains the real bot code with setup instructions.

Real usage

29%

Feedback from real users, and honest reflection on bugs.

Live handle + real build trail

29%

If the @handle doesn't respond when messaged, the demo link is dead, or the repo is a single copy-paste commit, this fails regardless of the write-up.

How we grade your AI usage

30% of your score

Using AI is the point — it’s the skill this certificate proves. You’ll answer three short questions about how you used it: what you asked, what was wrong with its first answer, and what you changed. Specific, honest answers score high. “I pasted the brief and submitted the answer” scores near zero.