Tagged “llm”

OpenAI's Decisions API Puts a Typed Endpoint on gpt-6-luna

10 Oct 2026

OpenAI's Decisions API returns predicates, choices and scores instead of free text. What's confirmed about it, how it differs from Jev under the hood and what I'd do about it in a production system.

3 min read·

llm agents inference architecture production

Jev Only Picks From the Answers You Give It

23 Sept 2026

TypeSafe AI's Jev returns typed answers with calibrated probabilities to a fixed set of questions and generates no tokens. How it works and where I'd put it in a production system.

2 min read·

llm agents inference architecture production

Ground Truth and LLM Judges Measure Different Things

28 Jul 2026

Why my thesis evaluation pipeline scored every run with both a deterministic benchmark and an LLM judge. The disagreements between them turned out to be the useful part.

2 min read·

evaluation llm rag architecture production

Using the LLM as a Query Planner

14 Jul 2026

In my thesis work the model only routes questions to the right data partition and pushes predicates down to PostGIS and Neo4j. Inference token cost dropped 3x and routing-index quality ended up deciding answer accuracy.

2 min read·

llm rag architecture query-planning production

Memvid Put a Vector Database in an MP4, Then Dropped the Video

19 Apr 2026

Memvid started as a joke about hiding AI memory inside MP4 video files. The serious version that replaced it drops the video entirely. It's worth explaining why.

3 min read·

rag ai-infrastructure vector-databases open-source llm

What Actually Happens When a RAG Pipeline Fails

18 Apr 2026

A field guide to the failure modes that don't show up in tutorials and how to systematically debug them.

3 min read·

rag debugging llm information-retrieval production

RAG vs Fine-Tuning: When to Use Which

16 Apr 2026

Everyone asks this question. Most of the answers online are wrong or at least incomplete.

3 min read·

rag fine-tuning llm machine-learning architecture