Most RAG Projects Need SQLite Before Vectors
A practical RAG ladder: start with SQLite FTS5, measure misses, then add embeddings only when semantic search earns its complexity.
NESTFRONTIER / ARCHIVE
All entries
Technical analysis, model releases, research, and deployment notes.
PAGE_07
346 PUBLISHED ENTRIESA 35B Agent Team Still Needs a Hard Stop
Apodex 1.1 shows when a 35B agent team earns its overhead, and when it only creates more plausible debris. Here is the safe local test path.
Always-On Agents Need a Budget and a Sandbox
Headlong makes AI agents think continuously. Here is the safe way to test one without an uncapped bill, exposed host, or shared-memory surprise.
A parser bug can turn local LLMs into host breaches
A real vLLM code-execution advisory changes local LLM security: isolate the inference host before model output reaches parsers, tools, or your network.
Coding skills lost 47 points to a short AGENTS.md file
Vercel found a short AGENTS.md docs index reached 100% in its evals, while default skills stayed at 53%. Here is when to use each approach.
128K Context Now Demands Hopper Hardware
FlashPrefill V2 reports 47.26x faster 128K FP8 prefill on H20. Here is the Hopper-only SGLang test path and the benchmark caveat.
Your Local LLM May Be Misconfigured, Not Dumb
Before downloading a larger checkpoint, debug your local LLM runtime. A wrong template, tiny context window, or bad sampler can make good weights look broken.
Live Code Is Powerful Until Your Agent Can Rewrite It
Autolith v0.35.0 can mutate a live Lisp runtime and recover after crashes. Here is the safe install boundary and the failure test worth running first.
Before you run parallel AI agents, add a local supervisor
Munder Difflin turns existing AI terminal CLIs into a local agent office. Here is the setup, the real safety boundary, and when the extra machinery is worth it.
Real-Time TTS Still Breaks at Six Requests Per Second
Nari Labs shows why real-time TTS is a scheduling problem: tuned engines can collapse under load even when one-request latency looks excellent.