Practical deep dives into AI applications, backend architecture, and open-source productivity tools.
LATEST POSTS
Anthropic removed over 80% of Claude Code's system prompt for Claude 5 — and got better results. Here's what the new rules of context engineering mean for anyone building with LLMs in 2026.
If you've ever read something written by an LLM and thought 'this feels… off,' you've stumbled on the very problem Humanizer was built to solve. Here's why this unassuming GitHub repo might be the most important agent skill in your toolkit.
Browser automation tools have often forced you to choose between keeping your tabs and letting the agent drive. Ego lite is a browser designed for parallel human-AI work, with separate Spaces for agents and people.
After years of wrestling with Conda environments and Python package conflicts just to run a few audio models, audio.cpp arrives with a single C++ runtime that reports faster results in some benchmarks and lower latency in selected workloads, while supporting everything from TTS to diarization.
Pydantic AI brings the same joy and type-safe ergonomics that FastAPI brought to web development—finally, an agent framework that doesn't make you want to tear your hair out.
Graphify turns any folder of code, docs, images, and videos into a queryable knowledge graph you can ask questions about. It's not a vector store — it's a real graph with deterministic, local-first code parsing. Here's why you should start using it today.
Prompt engineering is dead. The new leverage point is designing loops that orchestrate coding agents autonomously. This repo by Cobus Greyling shows you how.
Most AI courses teach scattered pieces. This one—503 lessons, 20 phases, four languages—builds everything from math to autonomous swarms. Every lesson ends with a reusable tool you can run today.
A practical guide to using Codebase Memory MCP — why it matters, how it works, and the surprising ways it reshapes your relationship with AI coding assistants.
Python 3.14 isn’t just another incremental release — it fundamentally rewrites how we handle annotations, strings, and concurrency. Here’s what you need to know about the three game-changers that landed in October 2025.
WRITING ABOUT
LLM application architecture, RAG system design & optimization.
Python, FastAPI, PostgreSQL, Redis, Celery, Kafka — backend patterns.
Open-source contributions, custom toolchains, and automation.
STAY IN TOUCH
Be the first to know when new articles are published.