Python Development Internship

Write Python that other people can run, read and trust.

The gap between "my script works" and "this is production software" is tests, structure, error handling and packaging. This programme closes it, using AI tooling the way working engineers now actually use it.

Beginner-friendly4 weeks · remote4 briefs · complete any 2Free to join

What you'll be able to do

Structure a Python project so a stranger can run it
Write tests that catch real regressions
Build and document a REST API
Set up CI that actually blocks bad merges

Who it suits

Beginners who can write Python scripts and want to learn how software is really shipped.

What you'll work with

Python
FastAPI
pytest
SQLite/Postgres
Docker
GitHub Actions

The project briefs

Complete any 2 of these 4to finish the programme. Assessment criteria are published here so you know what “good” means before you start.

CLI Tool with Real Error Handling

~8h estimated

Build a command-line tool that behaves well when things go wrong.

Pick a genuinely useful task. Then spend most of your time on the unhappy paths — missing files, bad input, network failure — because that is what separates a script from a tool.

Assessed against

Helpful error messages, never a raw traceback
Sensible exit codes
--help that explains itself
Handles being interrupted mid-run

You'll finish with

GitHub repo
README with install and usage

REST API with Tests

~12h estimated

Build a small API with a test suite you would trust in CI.

FastAPI with typed request/response models. Write tests for the failure cases first — the success path is the easy one and rarely where bugs live.

Assessed against

Four or more endpoints with typed models
Tests covering validation failures and edge cases
Auto-generated API documentation
A sane response for a malformed request

You'll finish with

GitHub repo
Test coverage report

Automation Script with Scheduling

~8h estimated

Automate a recurring task and make it safe to run unattended.

Anything unattended needs logging, idempotency and a plan for partial failure. Running it twice must not double the work.

Assessed against

Idempotent — safe to run twice
Structured logging
Configuration outside the code, secrets outside the repo
A documented recovery path when a run fails halfway

You'll finish with

GitHub repo
Sample logs from a real run

Package It and Ship It

~10h estimated

Turn one of your earlier tasks into something another person can install and run.

Dockerise it, add CI, write documentation aimed at someone who has never seen the project. Then have somebody actually try it — their confusion is the real test result.

Assessed against

Dockerfile that builds cleanly from scratch
CI running tests on every push
Documentation a stranger can follow
Verified by someone else actually running it

You'll finish with

GitHub repo with CI badge
Note on what confused your tester

Free to apply and free to complete. The completion certificate is optional.

Other domains

AI Agents & Automation

Build autonomous agents that use tools, not chatbots that answer questions.

Generative AI & LLMs

Go past prompt tinkering into retrieval, evaluation and cost control.

Machine Learning

Classical ML done properly — including the evaluation everyone skips.

Ready to Transform Your Enterprise with AI?

Book a free 30-minute consultation with our AI architects. Discover how agentic AI can automate your workflows and multiply your team's output.

No commitment required · Free architecture review · NDA protected

Python Development Internship

Write Python that other people can run, read and trust.

Beginner-friendly4 weeks · remote4 briefs · complete any 2Free to join

The project briefs

Complete any 2 of these 4to finish the programme. Assessment criteria are published here so you know what “good” means before you start.

CLI Tool with Real Error Handling

~8h estimated

Build a command-line tool that behaves well when things go wrong.

Pick a genuinely useful task. Then spend most of your time on the unhappy paths — missing files, bad input, network failure — because that is what separates a script from a tool.

Assessed against

Helpful error messages, never a raw traceback
Sensible exit codes
--help that explains itself
Handles being interrupted mid-run

You'll finish with

GitHub repo
README with install and usage

REST API with Tests

~12h estimated

Build a small API with a test suite you would trust in CI.

FastAPI with typed request/response models. Write tests for the failure cases first — the success path is the easy one and rarely where bugs live.

Assessed against

Four or more endpoints with typed models
Tests covering validation failures and edge cases
Auto-generated API documentation
A sane response for a malformed request

You'll finish with

GitHub repo
Test coverage report

Automation Script with Scheduling

~8h estimated

Automate a recurring task and make it safe to run unattended.

Anything unattended needs logging, idempotency and a plan for partial failure. Running it twice must not double the work.

Assessed against

Idempotent — safe to run twice
Structured logging
Configuration outside the code, secrets outside the repo
A documented recovery path when a run fails halfway

You'll finish with

GitHub repo
Sample logs from a real run

Package It and Ship It

~10h estimated

Turn one of your earlier tasks into something another person can install and run.

Dockerise it, add CI, write documentation aimed at someone who has never seen the project. Then have somebody actually try it — their confusion is the real test result.

Assessed against

Dockerfile that builds cleanly from scratch
CI running tests on every push
Documentation a stranger can follow
Verified by someone else actually running it

You'll finish with

GitHub repo with CI badge
Note on what confused your tester

Free to apply and free to complete. The completion certificate is optional.