Log in
New: A no-code editor for your ML projects

Watch Jams, build AI and ML projects, and get feedback from the community. Take your first line of code all the way to a project on GitHub.

Start learning How it works

WATCH. BUILD. SHARE.

Featured by the community

SPEKAWork.uaSTUD-POINT

WHO IS IT FOR?

PITMAPLUS
Pixel robot using a laptop
Build your own AI tools
DEVELOPMENT
01/ 04

FOR DEVELOPERS

Add AI and ML tools to your stack and go from using existing models to building solutions of your own.

PythonNeural NetworksApplied ML
DevelopmentAI Engineering
PITMAPLUS
From data to models
DATA SCIENCE
02/ 04

FOR DATA PROFESSIONALS

Build on your analytics and data science skills with machine learning, neural networks and modern AI systems.

Data ScienceMachine LearningAI Systems
AnalyticsMachine Learning
PITMAPLUS
Pixel-art cat
Getting started in AI & ML
START IN AI
03/ 04

NEW TO AI & ML?

Build a foundation in AI and ML with a sequence of lessons that guides you from one skill to the next.

PythonNumPyML Basics
FundamentalsYour first ML model
PITMAPLUS
Go further with AI
ADVANCED AI
04/ 04

READY TO GO BEYOND THE BASICS?

Explore RAG, reinforcement learning and other specialist topics that introductory courses rarely cover.

RAGReinforcement LearningLLM
AI basicsSpecialization

PITMA IN THE PRESS

LEARN AND PRACTICE / PITMA HUB

Learn ML.
Put it into practice.

Watch video lessons, build projects and get answers from the community, all in one place for free.

01 / VIDEO LESSONS

Learn ML
with our video lessons, called Jams.

Pick a topic, watch the explanation and follow the code, from Python basics to machine learning.

Watch Jams
PITMA
JamsProjectsForum
P
JAM LIBRARY

Python and ML video lessons

Find a lesson…
Choose a topic and watch a lessonPYTHON / AI / ML
Video lesson
Introduction to Python

Python for data science

Video lesson · 54 min
Create a project
Video lesson
Introduction to machine learning

AI and machine learning

Video lesson · 62 min
Create a project
Video lesson
Clustering algorithms

AI and machine learning

Video lesson · 41 min
Create a project

02 / YOUR PROJECTS

Build ML projects.
Share your work.

Start with an idea from a Jam, write the code and open a pull request. Use the review to improve your solution, then share it with the community and on GitHub.

Create a project
PITMA
JamsProjectsForum
P
Your project /

Implementing a transformer

Merged
Task and code Pull request Materials
attention.py attention → main
01def attention(q, k, v):
02 d = q.size(-1)
03 scores = q @ k.transpose(-2, -1)
04+ scores = scores / math.sqrt(d)
05 weights = scores.softmax(dim=-1)
06 return weights @ v
Changes saved in v2+1
PULL REQUESTCode ready for review

See the implementation and its changes together.

COMMUNITY REVIEWChanges requested

Add scaling before softmax.

APPROVEDChanges approved

Ready to merge into main.

Merged into main

Your project is now in the community gallery.

Write codeSubmit a PRGet feedbackPublish

03 / COMMUNITY

Forum
for your questions.

Stuck on code or a concept? Ask the community, talk through your approach and help others learn.

Join the community
PITMA
JamsProjectsForum
P
PITMA COMMUNITY

Forum

Ask a question or share a solution.

NewestHelpfulUnsolved
Search discussions…
PCommunity memberQuestion
How can I tell if my model is overfitting?

My model does well on the training data but much worse on new data. What should I check first?

Machine learning
MCommunity answer

Compare training and validation loss. If training loss keeps falling while validation loss rises, your model may be overfitting.

train_loss ↓val_loss ↑
AI LAB BETA

Build ML models.
Block by block.

Build a workflow without code. Choose a dataset, prepare the data and configure a model.
Open the Python notebook in Google Colab and check the results.

Open AI Lab
Workflow Results Python
EXAMPLE EXPERIMENT
Data

Customer churn

Task
Classification
Dataset
Customer churn
tenurechargesChurn
01 / DATASET
Preparation

25% held out for testing

Test size 25%
Missing values
Median
Scaling
StandardScaler
02 / TRAIN → TEST
Model

Logistic Regression

Algorithm
Logistic Regression
Regularization · C 1
Fit on training data
03 / TRAINING
Metrics

Accuracy · F1

Accuracy
Precision
F1 score

Evaluate the model after running the notebook in Colab.

04 / EVALUATION
Connect the blocks, set the parameters and run the experiment.
01 / DATA

Start with a preset.

Choose a regression, classification or clustering preset and adapt it to your task.

RegressionClassificationClustering
02 / EXPERIMENT

Compare models and metrics.

Try a different model or change a parameter. Run the notebook and import the result to see metrics and charts.

03 / PYTHON

From blocks to code.

Turn the workflow into a Python notebook. Open it in Google Colab, run the cells and follow each step.

model.fit(X_train, y_train)

Run your first ML experiment in your browser.

Try it for free

PITMA GUIDES

PITMA BLOG

All guides