Notebook
Writing, research, and talks.
A single place for Medium posts, local essays, applied research notes, and speaking ideas. Public enough to be useful, careful enough to protect private work.
April 18, 2026
Building a Persistent Knowledge Base for Claude Code: From Session Amnesia to Compounding Intelligence
A long-form engineering deep-dive into applying Karpathy's LLM-wiki pattern to Claude Code: transcript-replay auto-updates, Obsidian as source of truth, the honest trade-offs of Graphiti and custom builds, and a full survey of where else a personal knowledge base can live in 2026.
January 27, 2026
AI Receipt Splitter: Privacy-First Bill Splitting with Claude Vision
A technical deep-dive into building a stateless, multi-currency receipt splitter using Claude Vision OCR, Go Fiber, and React—with zero user data storage.
January 12, 2026
Quantum-Proof Password Generator: Implementing NIST FIPS 203
A practical implementation of a Post-Quantum Credential Lifecycle system using NIST FIPS 203 (ML-KEM) and Argon2id on AWS Lambda.
December 19, 2025
Why I Dumped Node.js for Bun (And You Should Too)
It's not just hype. Migrating this website to Bun cut my install times by 90% and made the dev server instant. Here's the data.
December 18, 2025
Building Scalable Systems with AWS SAM
A deep dive into serverless architecture, why I moved away from Argo CD, and how to optimize for cost and performance.
Medium
Latest from Medium
Feb 24, 2026
The Quiet Tax of Over-Engineering: Why “Boring” Tech Wins
There’s a particular kind of developer — sharp, curious, always reading Hacker News — who rewrites a working cron job in Rust because it could be faster, or reaches for Kafka because a YouTube talk made pub/sub sound exc
Jan 12, 2026
I Built a Quantum-Proof Password Generator — Here’s Why You Should Care
I Built a Quantum-Proof Password Generator — Here’s Why You Should Care Post-Quantum Cryptography isn’t science fiction anymore. NIST has standardized it, Go 1.24 has implemented it, and I deployed it to AWS Lambda. Here
Nov 24, 2025
Sentinel Alert App: Real-Time Incident Monitoring & Response, Built with Go, Redis, Postgres & PWA
Overview Sentinel Alert App is a real‑time incident monitoring, alert streaming, and on‑call dashboard built using Go , PostgreSQL , Redis , Docker , and a PWA frontend. This project was created to provide a lightweight,
Jul 27, 2025
Building My Own Local LLM Chatbot with Docker Model Runner
🤔 Why Build Your Own LLM Chatbot? There are already countless AI chatbots online — from ChatGPT to Claude to Perplexity. So why did I decide to build my own? Because I believe that exploring and building is more valuabl
Jul 25, 2025
Running Docker Model LLMs in CI/CD for Automated Code Review
“The power of the LLM in the palm of your CI pipeline.” Introduction LLMs are transforming how we code — but using them in your CI/CD pipeline for automated code review? Now that’s next-level productivity. In this post,
Jul 13, 2025
Docker Model Run: The Simplest Way to Run Open-Source AI Models Locally
Tired of API rate limits and OpenAI billing surprises? With docker model run, you can run AI models like Mistral, LLaMA, or even Stable Diffusion locally — with just one command . Introduction Over the past year, the ris
Jul 5, 2025
Structuring Laravel APIs for Enterprise Projects: My Controller–Service–Resource Pattern
When building APIs in Laravel for production-ready, enterprise-scale applications, structure matters — a lot. I’ve worked on several Laravel projects where keeping the logic clean, modular, and testable wasn’t just a nic
Jun 28, 2025
Why I Moved My Personal Website to Cloudflare Pages (And Why You Might Want To)
Like many developers, I initially hosted my personal site on Vercel using Next.js . It worked well — great DX, fast deploys, clean UI. But over time, I started hitting limits, especially around cold starts and bandwidth
Research
Applied notes you can open
AI systems / reading
AI-Assisted Operations Without Losing Control
A reading and prototype trail around alert summarization, incident triage, and private engineering assistants.
Applied security / prototype
Post-Quantum Cryptography for Everyday Web Apps
Notes on practical ML-KEM demos, hybrid transport, and what developers should learn before quantum-safe becomes default plumbing.
Talks