iiinova — where your digital self keeps evolving

iiinova is part of the Atmosphere — an open, public network where you own your identity.

Pau Labarta Bajo

@paulabartabajo.com

329 followers 20 following 146 posts

Teaching Real World ML at realworldml.net - ML Freelance engineer - Maths Olympian - Father of 1.. sorry 2 kids

Posts Replies Media

Pau Labarta Bajo @paulabartabajo.com · 06/02/2025

Ever felt like this?

021 2

Pau Labarta Bajo @paulabartabajo.com · 01/02/2025

There is (very) intelligent life beyond LLMs...

010 1

Pau Labarta Bajo @paulabartabajo.com · 26/01/2025

Without data engineering... ... there is no LLM engineering.

130 3

Pau Labarta Bajo @paulabartabajo.com · 16/01/2025

Today is the first day of the rest of your life. Choose your path wisely.

010 1

Pau Labarta Bajo @paulabartabajo.com · 30/12/2024

How do you monitor your ML models?

000 0

Pau Labarta Bajo @paulabartabajo.com · 22/12/2024

Rate my setup

010 1

Pau Labarta Bajo @paulabartabajo.com · 16/12/2024

→ Inference pipelines, that load the model from the registry, and the input from the client app (for example a vector of numerical features, or a text prompt), generate a prediction (or a generation) and return it to the client app.

100 1

Pau Labarta Bajo @paulabartabajo.com · 16/12/2024

→ Training (or fine-tuning) pipelines, that read historical features from the Feature Store/Vector DB and generate a new model artifact, either by training from scratch or fine-tuning a base LLM. This model artifact is then pushed to the model registry.

100 1

Pau Labarta Bajo @paulabartabajo.com · 16/12/2024

Any ML system can be decomposed into at least 3 types of programs (aka pipelines) → Feature pipelines, that transforms raw data into ML model features (e.g. vector embeddings) that are saved in a Feature Store or Vector DB.

110 1

Pau Labarta Bajo @paulabartabajo.com · 12/12/2024

Don't now how to start an ML project? This is the way ⬇️

020 2

Pau Labarta Bajo @paulabartabajo.com · 27/11/2024

4 years ago I struggled to land my first freelance ML engineering contract. Then I discovered this ↓

100 1

Pau Labarta Bajo @paulabartabajo.com · 25/11/2024

Deploying ML models is 𝗲𝗮𝘀𝘆... ... when you use the 𝗿𝗶𝗴𝗵𝘁 𝘁𝗼𝗼𝗹𝘀

010 1

Pau Labarta Bajo @paulabartabajo.com · 23/11/2024

You choose

030 3

Pau Labarta Bajo @paulabartabajo.com · 23/11/2024

Advice for ML beginners 💡 → A machine learning model is a product of the dataset it was trained on. → Bad data means a bad model. → Good model means a good dataset.

030 3

Pau Labarta Bajo @paulabartabajo.com · 22/11/2024

The First Rule of Machine Learning ⬇️

031 3

Pau Labarta Bajo @paulabartabajo.com · 21/11/2024

Claude aged really well

010 1