Technical Writings

Thoughts on AI, machine learning engineering, and system optimization.
Synced dynamically from Medium

Jan 21, 2026

Beyond print(): Building Production-Grade Observability in Google ADK

In traditional software development, debugging is a deterministic process. If a function fails, you check the inputs, trace the logic, and fix the bug. AI Agents are different. They are probabilistic engines that think,...

Read full story
Dec 16, 2025

Quasi-Random Search: A Smarter Way to Tune Hyperparameters in Python

We’ve all been there. You kick off a Grid Search on Friday afternoon, hoping to come back to a perfectly tuned model on Monday. Instead, you return to find that 90% of the computation was wasted checking irrelevant...

Read full story
Sep 19, 2023

TensorFlow’s Guide to Fine-Tuning BERT

In today’s ever-evolving world of Natural Language Processing, BERT (Bidirectional Encoder Representations from Transformers) is the foundation for many subsequent models. With its pre-trained contextual language...

Read full story
Aug 6, 2023

Wiping Out Git History and Creating a Fresh Branch

There are times in the lifecycle of a software project where starting fresh is desirable or even necessary. This article delves into one such scenario — deleting all the commits from your Git repository and creating a...

Read full story
Aug 4, 2023

Unraveling Spotify’s Music Universe: A Clustering Analysis Approach

Ever wondered how music recommendation systems suggest songs that align perfectly with your taste? The secret lies in the data-driven techniques used to analyze, classify, and cluster millions of songs. This article...

Read full story