The 2026 AI Hacker Stack
After fine-tuning 40+ models this year, here's the stack that compounds.
Long-form pieces on the operational decisions that actually matter in 2026.
After fine-tuning 40+ models this year, here's the stack that compounds.
LangChain, CrewAI, AutoGen — why the 200-line custom loop wins.
Red-team research on Claude, GPT-5, Gemini — the patterns that still work.
After deploying RAG at 6 companies, here's what works and what's theater.
Gemma 3 27B on a 4090. The local-AI playbook that ships.
200 lines of custom code beats the framework. Every time.
Most 'prompt engineering' is bullshit. The 5 patterns that actually move metrics.
LoRA + base models + small specialists. The 2026 fine-tuning playbook.
Three months of production benchmarks. The honest comparison.
Why we run our own evals, train our own models, and don't trust the hype.