Tag
#self-hosting
Every story tagged self-hosting, newest first.
Open-Weight Models Changed Who Controls the AI Stack
Open-weight models are now good enough that many teams no longer depend on a single vendor's API. The real shift isn't cost: it's control and portability. The same weights can run in-house or across competing hosts, so no one provider owns your stack by default.
Ahmad J · Aug 12, 2026 · 3 min read
What a Model Actually Costs to Run in Production: A Back-of-Envelope Framework for Teams
Training makes headlines; inference quietly compounds every month until the cloud bill arrives. A reusable back-of-envelope framework (tokens, caching, batch tiers, latency, and engineering time) to estimate what a deployed model really costs before you commit.
Ahmad J · Aug 9, 2026 · 8 min read
The Real Privacy Audit: What Data Your AI Coding Assistant Sends Home
Stop trusting privacy policies. Put your AI coding assistant behind a proxy and watch exactly what your code transmits: what fires on every keystroke, what "local mode" really hides, and how to shut the channel.
Ahmad J · Aug 5, 2026 · 11 min read
The best vector databases for RAG in 2026
Qdrant is still the safe default for new RAG projects in 2026, but the field moved on cost: Milvus 2.6 shipped 1-bit quantization, Pinecone launched Nexus to public preview, and object-storage-first newcomers Turbopuffer and LanceDB undercut the incumbents. An honest buyer's guide.
Ahmad J · Aug 1, 2026 · 12 min read
The best AI image generators in 2026
GPT Image 2 is the best all-round AI image generator in 2026, but not the only story. Google's Nano Banana Pro, Riverflow 2.0 Pro, Midjourney V8.1, Firefly 5, Ideogram 3.0 and the FLUX.2 family each win a different job. Here's who should pick what, with rough mid-2026 pricing.
Ahmad J · Aug 1, 2026 · 13 min read