Imad Dabbura

Hard-Learned Lessons in Shipping Software (AI/ML) Projects

A Guide for Engineers and Product Managers

Product Management

Jan 5, 2025

Imad Dabbura

5 min

Jan 5, 2025

From Forgetting to Fluency: How to Learn Smarter, Not Harder

Evidence-based tips from cognitive science to transform the way you absorb, retain, and apply knowledge in any field.

Personal Growth

Why You Should Ditch Softmax in Your Neural Network’s Final Layer

Discover the hidden drawbacks of softmax—and why raw logits often lead to faster, more stable, and more accurate training

Deep Learning

Cutting the Fat: A Practical Guide to Neural Network Pruning

From weight sparsity to hardware-aware strategies—learn how to shrink deep learning models without shrinking their performance

Deep Learning

Building GPT(2/3) from Scratch: Turning Theory into a Working Transformer

A hands-on journey through implementing the 124M GPT architecture in PyTorch—complete with attention mechanisms, optimizations, and the lessons learned along the way

NLP

Deep Learning

Tokenization Uncovered: How BPE Shapes the Mind of a Language Model

A deep dive into the algorithm that compresses text, balances trade-offs, and defines how AI perceives meaning

NLP

Deep Learning

The RAG Optimization Playbook

Proven tactics, pitfalls, and fine-tuning methods to build faster, smarter, and more accurate retrieval-augmented generation systems

NLP

RAG

Inside Python’s Modules and Packages: The Machinery Behind import

Explore the inner workings of Python’s import system, including path resolution, package types, importlib internals, and advanced import hacks

Python

SWE

Automatic Differentiation Demystified

From dual numbers to backpropagation, uncover the inner workings of forward and reverse mode AD and the performance trade-offs that shape deep learning

MLSys

Designing ML Systems

MLSys

Git from the Inside Out

An in-depth guide to Git’s object model, branching mechanics, and the hidden workflows that make it so powerful

SWE

I Built My Own PyTorch (Tiny Version) — Here’s Everything I Learned

Inside the engineering decisions, optimizations, and trade-offs behind a homegrown deep learning framework

MLSys

Attention Is All You Need… But Here’s the Rest

A practical, code-first breakdown of Transformers—covering the theory, the math, and how to implement every architecture variant

NLP

Deep Learning

Breaking Text Apart (The Smart Way)

From single characters to advanced subword splits — see how modern tokenizers like WordPiece and SentencePiece prepare language for AI.

NLP

Inside LSTMs: Implementing and Optimizing Sequential Models from First Principles

A deep dive into LSTM internals—covering the math, gates, performance considerations, and a full PyTorch-aligned implementation from scratch.

NLP

Deep Learning

C Program Startup

Does C program really start at main

SWE

Airflow Part 8 - Best Practices

Data Engineering

MLOps

Airflow

Airflow Part 7 - Triggering Workflows

Data Engineering

MLOps

Airflow

Airflow Part 6 - Sharing Data Between Tasks

Data Engineering

MLOps

Airflow

Airflow Part 5 - Dependencies Between Tasks

Data Engineering

MLOps

Airflow

Airflow Part 4 - Task Context & Jinja Templating

Data Engineering

MLOps

Airflow

Airflow Part 3 - DAG Scheduling

Data Engineering

MLOps

Airflow

Airflow Part 2 - DAGs

Data Engineering

MLOps

Airflow

Airflow Part 1 - What is Airflow?

Data Engineering

MLOps

Airflow

Anomaly Detection

Machine Learning

Data Science

Conda Essentials

Enough background about Conda to be productive!

SWE

Gradient Descent Algorithm and Its Variants

Deep dive into gradient descent algorithm: Batch vs. Mini-batch vs. Stochastic.

Machine Learning

Deep Learning

K-means Clustering: Algorithm, Applications, Evaluation Methods, and Drawbacks

Deep dive into K-means algorithm to find subgroups within data.

Machine Learning

Data Science

Unsupervised Learning

Coding Neural Network Part 5 - Dropout

What is Dropout, its use, and how to implement it?

Machine Learning

Deep Learning

Coding Neural Network Part 4 - Regularization

What is regularization and how it helps NN generalizes better?

Machine Learning

Deep Learning

Coding Neural Network Part 3 - Parameters’ Initialization

The role of parameter initialization in training and different ways to initialize parameters.

Machine Learning

Deep Learning

Coding Neural Network Part 2 - Gradient Checking

How to check numerically if the implementation of backward propagation is correct?

Machine Learning

Deep Learning

Coding Neural Network Part 1 - Forward & Backward Propagation

What it takes to go from input to output? And how to compute the gradients?

Machine Learning

Deep Learning

Bandit Algorithms: epsilon-Greedy Algorithm

What is epsilon-Greedy Algorithm and how to use it in A/B testing?

Data Science

Website Optimization

Predicting Loan Repayment

Trying different modeling techniques to deal with imbalanced data, missing values, and ensemble models.

Machine Learning

Data Science

Character-Level Language Model

Predict the next character given the previous charecter and state.

NLP

Deep Learning

Predicting Employee Turnover

Experimenting with different models on employee turnover data.

Machine Learning

Data Science

Posts

Categories

Hard-Learned Lessons in Shipping Software (AI/ML) Projects

From Forgetting to Fluency: How to Learn Smarter, Not Harder

Why You Should Ditch Softmax in Your Neural Network’s Final Layer

Cutting the Fat: A Practical Guide to Neural Network Pruning

Building GPT(2/3) from Scratch: Turning Theory into a Working Transformer

Tokenization Uncovered: How BPE Shapes the Mind of a Language Model

The RAG Optimization Playbook

Inside Python’s Modules and Packages: The Machinery Behind import

Automatic Differentiation Demystified

Designing ML Systems

Git from the Inside Out

I Built My Own PyTorch (Tiny Version) — Here’s Everything I Learned

Attention Is All You Need… But Here’s the Rest

Breaking Text Apart (The Smart Way)

Inside LSTMs: Implementing and Optimizing Sequential Models from First Principles

C Program Startup

Airflow Part 8 - Best Practices

Airflow Part 7 - Triggering Workflows

Airflow Part 6 - Sharing Data Between Tasks

Airflow Part 5 - Dependencies Between Tasks

Airflow Part 4 - Task Context & Jinja Templating

Airflow Part 3 - DAG Scheduling

Airflow Part 2 - DAGs

Airflow Part 1 - What is Airflow?

Anomaly Detection

Conda Essentials

Gradient Descent Algorithm and Its Variants

K-means Clustering: Algorithm, Applications, Evaluation Methods, and Drawbacks

Coding Neural Network Part 5 - Dropout

Coding Neural Network Part 4 - Regularization

Coding Neural Network Part 3 - Parameters’ Initialization

Coding Neural Network Part 2 - Gradient Checking

Coding Neural Network Part 1 - Forward & Backward Propagation

Bandit Algorithms: epsilon-Greedy Algorithm

Predicting Loan Repayment

Character-Level Language Model

Predicting Employee Turnover