AI/ML Engineering & LLMOps

Training/inference, vector search, RAG, evaluation, safety, and production ML/LLM stacks.

  • 5 Subtopics
  • 14 Tracked terms
  • Last 30 days Feed window

Inside AI/ML Engineering & LLMOps

What this topic collects on

An article joins this feed when it matches these terms. Each one is also a search of its own.

Latest in AI/ML Engineering & LLMOps


dev.to > shaam_ai > glm-53-flash-explained-the-320b-open-weight-model-with-an-18b-brain-and-a-1m-token-memory-2026-5cbl

GLM-5.3-Flash Explained: The 320B Open-Weight Model With an 18B Brain and a 1M-Token Memory (2026)

49+ min ago   (651+ words) At a glance — Last verified: 2026-09-03 In one sentence: it is a 320-billion-parameter open-weight Mixture-of-Experts model from Z.ai that uses only 18 billion active parameters per token, reads text, images, video, and files natively, and holds a one-million-token conversation in a…...


dev.to > sindhuja_sudhakar > when-ai-agents-hammer-toolslist-building-a-caching-aware-mcp-server-sep-2549-38k6

When AI Agents Hammer tools/list: Building a Caching-Aware MCP Server (SEP-2549)

1+ hour, 4+ min ago   (555+ words) Part 2 of 2. Part 1 covered statelessness; here I tackle caching. This isn't hypothetical. When I pointed GitHub Copilot CLI at my server, it fired nine tools/list calls in ~40 seconds in one short session — plus repeated re-initializes. That's the storm SEP…...


dev.to > cheng2zhang > making-chroma-work-in-chinese-57jb

Making Chroma Work in Chinese

1+ hour, 29+ min ago   (1282+ words) Note: This tutorial picks up where Chroma's official quick start guide leaves off. So you've finished Chroma's quick start — congratulations. Everything works, right up until you feed Chroma some Chinese text. Chroma's default embedding model, all-MiniLM-L6-v2, is English-only. Given Chinese…...


dev.to > megapixel99 > a-determinism-check-has-to-leave-the-process-3ink

A Determinism Check Has to Leave the Process

1+ hour, 34+ min ago   (1536+ words) Code: Megapixel99/nondet The obvious way to check whether a Python function is deterministic is to call it twice and compare. I wrote that check, and it is blind to the commonest source of nondeterminism in the language. String hashing is…...


dev.to > atsushi_hatchco > a-new-model-dropped-dont-just-swap-the-id-2a3e

A New Model Dropped. Don't Just Swap the ID.

1+ hour, 46+ min ago   (845+ words) The new model is out. The benchmarks look great. Your provider has already scheduled the sunset date for the one you are running. So you swap the model ID, run a few requests, see nothing obviously broken, and ship. I…...


dev.to > ikkun1222 > jev-vs-a-310m-encoder-i-trained-myself-750-rows-three-tasks-two-different-winners-242e

Jev vs a 310M encoder I trained myself: 750 rows, three tasks, two different winners

2+ hour, 6+ min ago   (960+ words) One-line summary: A decision API wins when you have no labels; a small trained encoder wins when you have a few hundred — and which one wins on a given task is decided by the shape of the task, not by…...


dev.to > veil_im > we-label-every-security-claim-on-our-site-implemented-experimental-or-not-claimed-1jge

We Label Every Security Claim on Our Site: Implemented, Experimental, or Not Claimed

2+ hour, 5+ min ago   (497+ words) Somewhere in the last decade, "end-to-end encrypted" stopped meaning anything. So we did something that felt commercially reckless: we labelled every single claim on the site as one of three things. Implemented — it's in the current build, testable by anyone…...


dev.to > shahbaz_ali > one-predicate-two-meanings-three-bugs-building-a-contradiction-resolver-with-the-sanity-app-sdk-3jm7

One Predicate, Two Meanings, Three Bugs: Building a Contradiction Resolver with the Sanity App SDK

2+ hour, 14+ min ago   (1264+ words) by shahbaz_ali. Built for the DEV × Sanity Challenge. #sanitychallenge When two of your company's... Tagged with sanitychallenge, sanity, devchallenge, ai....


dev.to > heylittlepan > finding-the-right-terminal-in-a-room-full-of-agents-5986

Finding the right terminal in a room full of agents

3+ hour, 6+ min ago   (395+ words) I don't lose terminals when they're running. I lose them when I come back. An hour ago I had five... Tagged with terminal, ai, devtools....


dev.to > subaruhello > who-decides-the-tenant-a-small-rust-guard-for-ai-tool-calls-4pc4

Who Decides the Tenant? A Small Rust Guard for AI Tool Calls

3+ hour, 5+ min ago   (622+ words) I have been thinking about a fairly ordinary failure mode in AI-enabled SaaS products. A support... Tagged with rust, ai, security, showdev....