Sarah Glasmacher
Sarah Glasmacher
my journey through ML and studies
⌘K
Subscribe Sign in
ML Ops

🚀 Ready to Finally Upgrade from messy ML notebooks? Setup to Shipping Challenge

Tired of hard-to-maintain ML code? Join me on the 4-week ML Project Blueprint Challenge and turn a blank repo into a clean, configurable, and shippable ML project.
Sarah Glasmacher

by Sarah Glasmacher

There's a big difference between writing a data science notebook with sklearn pipelines and actually shipping ML code so it can run reliably and not on your laptop 😶 In the past 2 years in my job I have learned a lot about deployments in real life (not at FAANG) and will put them together in this short challenge format this month.

You can start anytime this month, go at your own pace, and by the end you’ll have:

  • A clean repo structure ready for any new ML project
  • Modular code with proper imports ready to be extended and tested
  • A config system using Pydantic (goodbye hard-coded paths)
  • Basic MLflow tracking for experiments
  • A Python package you can deploy to the cloud (Databricks example included)

Here's what we are doing this month (we, if you are joining me, otherwise just me 😔)

📆 4 weeks of short, weekly tasks (start anytime!)
✅ Get the previous week’s “solution” in each email
🏗️ End with a reusable ML project blueprint
💰 Free for newsletter readers this month

I’m doing this because I want a ready-to-use ML project blueprint for myself - something I can copy for every new proof of concept at work or in side projects. You’ll have the same: a starting point with all the best practices built in, that you can adjust in whatever way needed.

Week-by-Week Outline

Week 1: Your ML Repo Starter Kit: src Layout, uv venv, GitHub Push

Your Week 1 Task (Start Now) 💪🏻

Create the best version of a folder structure for a new ML project that you can think of - no code yet, just the scaffolding.

Need some ideas?

You can also use my guide on how I use uv, but that's already almost cheating ⚠️

Subscribe to receive new posts as email

No spam, no sharing to third party. Only you and me.

Member discussion

More like this

Your ML Repo Starter Kit: src Layout, uv venv, GitHub Push

Your ML Repo Starter Kit: src Layout, uv venv, GitHub Push

Let's start building an organized repository for our ML Blueprint Project 😊 This is the initial step in the
How I Set Up a Machine Learning Project with `uv` and `pyproject.toml`

How I Set Up a Machine Learning Project with `uv` and `pyproject.toml`

One of the biggest challenges in machine learning is ensuring that an experiment can be reliably reproduced. Of course, this
Using Taskfile with uv and pyproject.toml to manage your Python Machine Learning projects

Using Taskfile with uv and pyproject.toml to manage your Python Machine Learning projects

"Use uv to manage your Python projects" is not a new take anymore, but have you tried Taskfile?
“Side-by-side comparison of a disorganized ML project folder and a refactored version using the Python src/ layout”

Adding a src/ Folder to an Existing ML Project: Step-by-Step Guide

So you already created a repository for your ML project, but now you want a better way to organize your
What is MLflow model logging and when do I use which logging method?

What is MLflow model logging and when do I use which logging method?

If you’ve trained a machine learning model and thought “Wow, this one is actually good!”, only to realize a
How to set up pgvector with Docker: Local Vector Database for Text Embeddings

How to set up pgvector with Docker: Local Vector Database for Text Embeddings

In this post, I'll walk you through how I set up a local vector storage system using pgvector,