Archives
All the articles I've archived.
-
Fixing Codex + LiteLLM + Vertex AI: The Tool Types Conflict
How we resolved a 400 Bad Request error caused by non-standard tool types when proxying Codex requests to Vertex AI through a local LiteLLM proxy, ultimately patching LiteLLM's source code in our virtual environment.
-
Exporting Google ADK Agent Logs to BigQuery: A Complete Guide
Learn how to export your Google ADK agent execution logs to BigQuery for powerful analytics, debugging, and monitoring workflows.
-
The Insider's Guide to Building Enterprise RAG with Vertex AI RAG Engine
The definitive guide to building production-grade RAG systems with Google's Vertex AI RAG Engine, covering architecture, data ingestion, embedding models, vector databases, retrieval strategies, and complete working code. As of early 2026.
-
Managing Dual GitHub Identities Like an Insider
A guide to switching GitHub identities automatically on macOS using Git and the GitHub CLI.
-
Demystifying Python Classes with Real-World Examples
A friendly walkthrough of Python classes using concrete examples and everyday analogies.
-
Running Managed Open‑Source Models on VertexAI
A hands‑on guide to deploying managed open source models on VertexAI using LiteLLM proxy
-
Top Talent: The OpenAI Database Story
An analysis of OpenAI's database scaling and the CMU talent pipeline behind it.
-
When AI Meets Reality: Salesforce’s Layoff Saga, Deterministic Triggers, and the Legal Tightrope
A deep dive into Salesforce’s AI‑driven layoffs, the shift from large language models to deterministic automation, and the emerging legal risks of prompt‑based discounts in customer‑service bots.
-
Astro Joins Cloudflare: What It Means for Your Next Content‑Driven Site
Explore how Cloudflare’s acquisition of Astro reshapes open‑source governance, deployment, and AI opportunities for the fastest content‑first framework.
-
2026 Tech Landscape: Visa Fees, Quantum Edge, and SMR Power
The year 2026 is shaping itself as a crucible of change for the technology sector. Three forces are converging: a shift in U.S. immigration policy that raises…
-
Cutting MCP Token Costs: Search vs. Programmatic Tool Invocation
Explore how to slash token usage in Model Context Protocol (MCP) by leveraging on‑demand tool search or programmatic orchestration, with concrete examples from Claude and Cloudflare’s Code Mode.
-
NVIDIA Rubin: Six‑Chip AI Supercomputer and 10× Inference Token Cost Savings
Explore NVIDIA’s Rubin platform – a six‑chip rack‑scale supercomputer, its 10× lower inference token cost, and the observability tools that make large‑scale AI reliable.
-
Agent‑Browser: A Lightweight Headless‑Browser CLI for AI Agents
A quick guide to installing and using agent‑browser, a fast Rust‑based headless‑browser CLI that lets AI agents control Chromium from the terminal.
-
Creating User-Scoped Artifacts in Google ADK
Learn how to create user-scoped artifacts in Google ADK that persist across sessions, enabling better file management and user experience in AI agents.
-
Why I’m Building open-services-agent-starter-pack (Google ADK, without the cloud lock-in)
After building with LangGraph and Google ADK, I prefer ADK’s developer experience—but many core services assume Vertex AI. This WIP starter pack is my attempt to make ADK feel ‘first-class’ on non-GCP infrastructure.
-
Self-Hosting Photos with Immich: An Architectural Deep Dive
A deep dive into self-hosting your photos and videos with Immich, exploring its architecture, features like semantic search and facial recognition, and backup strategies.
-
HP Pavilion Battery Issues
A personal story about troubleshooting a persistent battery drain issue on an old HP Pavilion laptop, diving into hardware upgrades and OS diagnostics.
-
Architecting Analytics Backend
This blog post details the architectural decisions and philosophies behind building a robust backend system for a startup, focusing on FastAPI, PostgreSQL, and advanced SQL techniques.
-
Architecting a Retention & Growth Platform Backend
This blog post details the architectural decisions and philosophies behind building a robust backend system for a retention and growth platform startup, focusing on FastAPI, PostgreSQL, and a modular design.
-
Maximizing the BigQuery Free Tier: A Complete Cost Control Guide
Learn how to maximize Google BigQuery's free tier while avoiding unexpected costs. Discover smart querying strategies, data modeling techniques, and essential cost control measures.
-
Pandas vs Polars vs DuckDB - Performance Showdown on Flight Data
A comprehensive performance comparison between Pandas, Polars, and DuckDB using a large flight delay dataset. Discover which library reigns supreme for data analytics.
-
OpenAI Just Dropped a Bombshell: Meet gpt-oss, Their New Open-Weight Models
OpenAI has released gpt-oss-120b and gpt-oss-20b as open-weight models under Apache 2.0 license. Discover their impressive benchmarks, efficient MoE architecture, and how they compare to proprietary models.
-
A Deployment Story: Taking a FastAPI AI Chatbot Live on DigitalOcean
How to deploy your fastapi application to Digital Ocean Droplet
-
Internals of Langgraph Postgres Checkpointer (AsyncPostgresSaver)
An in-depth developer's guide to LangGraph's PostgreSQL checkpointer, its schema, and practical strategies for high-performance state management in LLM applications.
-
Fine-Tuning Phi3-mini for coronary artery disease Prediction
Learn how to fine-tune Microsoft's Phi-3-mini model for heart disease prediction using LoRA and quantization techniques. A complete guide from setup to deployment.
-
How to Build and Deploy a LangGraph Chatbot with FastAPI - Complete Guide (Part 1)
A comprehensive guide showing how to create a stateless chatbot using LangGraph and deploy it as a production-ready API using FastAPI
-
LangGraph Core Concepts A Simple Explainer
A short and concise explanation of the core concepts of LangGraph. It is designed to be easily understandable for anyone new to the framework.
-
Meet Kimi K2 – A Chill Intro to Agentic AI
Discover Kimi K2, the cutting-edge agentic AI model with 1 trillion parameters. Learn about its tool-using capabilities, impressive benchmarks, and how to integrate it into your projects today.
-
How to add Supabase OAuth in FastAPI
A comprehensive guide on integrating Supabase OAuth with FastAPI for secure social logins using GitHub, Google, and other providers.
-
Barriers, Giants, and Imaging Wars: The Competitive Moats of Medical Imaging
Exploring barriers, giants, and imaging wars: the competitive moats of medical imaging
-
Essays on banking: is debt at arms-length always a good option for your business?
Exploring essays on banking: is debt at arms-length always a good option for your business?
-
Transforming Clinical Trials: Adaptive Approaches for Better Patient Outcomes
Exploring transforming clinical trials: adaptive approaches for better patient outcomes
-
On Singular value decomposition and it's use in recommendation systems
Exploring on singular value decomposition and it's use in recommendation systems
-
Economics of Savage weddings
Exploring economics of savage weddings
-
Irrational to Rational
Exploring irrational to rational