//Writing

Blog

Notes on AI/ML, large language models, prompt engineering, and building software that ships. Written from hands-on experience.

Jul 22, 20263 min read

Practical Python Performance Optimization for AI Workloads

Learn how to optimize Python performance for AI workloads, covering topics such as just-in-time compilation, caching, and parallel processing. This article provides practical guidance on improving the performance of your Python-based AI applications.

Python PerformanceAI WorkloadsOptimization

Jul 20, 20264 min read

Building a Minimal LLM Agent Tool Loop in Python

A practical guide to building a minimal LLM agent tool use loop in Python without heavy frameworks, with runnable code.

AI AgentsPythonTool Use

Jul 20, 20263 min read

Efficient Model Deployment with FastAPI and Kubernetes

Learn how to build a scalable model deployment pipeline using FastAPI and Kubernetes. This article covers the design and implementation of a high-performance model serving system.

MLOpsFastAPIKubernetes

Jul 16, 20263 min read

RAG and Beyond: Advanced Question Answering with Retrieval-Augmented Generation

This post explores the applications and implementation of Retrieval-Augmented Generation (RAG) models in advanced question answering systems, including model training and deployment using FastAPI and PyTorch. We discuss RAG model architecture and its potential in various applications.

RAGQuestion AnsweringPyTorch

Jul 13, 20264 min read

Designing Efficient Vector Search Systems with Python and Faiss

This article explores the design and implementation of efficient vector search systems using Python and the Faiss library. We will discuss the key concepts and provide a practical example of building a vector search system.

Vector SearchFaissPython

Jul 10, 20263 min read

Optimizing React Performance with Next.js and Rust

Learn how to improve React application performance using Next.js and Rust. This guide covers optimization techniques and best practices for building high-performance React applications.

ReactNext.jsRust

Jul 9, 20263 min read

Building Scalable MLOps Pipelines with FastAPI and Docker

Learn how to deploy machine learning models using FastAPI and Docker, focusing on scalability and maintainability. This post covers the design of MLOps pipelines and best practices for model deployment.

MLOpsFastAPIDocker

Jul 8, 20263 min read

Fine-Tuning LLMs on a Budget: A Practical Guide

How to fine-tune large language models without a GPU cluster, using LoRA, quantization, and a single consumer GPU. A hands-on walkthrough with code.

LLMPythonMachine Learning

Jul 7, 20262 min read

Prompt Engineering Patterns That Actually Work

Beyond 'be concise', the prompt patterns I reach for in production LLM apps: few-shot scaffolding, structured output, and self-checking chains.

LLMPrompt EngineeringAI