Two free deep-dives: multimodal RAG bottlenecks in order and shipping agents in a real product
ubiquitous_tech · reddit · 2026-10-28
A practitioner posted two free long-form video breakdowns arguing that most production pain is boring infrastructure, not prompt wording.
- Multimodal RAG bottlenecks in order: parsing → chunking → embedding → retrieval → reranking → generation, flagging where hello-world tutorials silently fail (PDF tables, scanned pages, multilingual, multi-vector).
- Agents in a real product: mental model of LLM + tools + memory + stop conditions, structured outputs/tool calls, evals before multi-agent, and common production failure modes.
Suggested first exercise: one agent, three tools (search, calculator, local file), measured pass/fail on 10 fixed scenarios before adding RAG. Newcomers are pointed to Andrej Karpathy's intro material first.
More from coding & agent
- Claude Catches North Korean Supply Chain Attack Disguised as an Auth Error — banteg · 2026-10-28
- Expo ships agent skills collection for AI coding tools, already 3.3k GitHub stars — expo · 2026-10-28
- Does a Google Play uploader MCP exist, and would you trust an agent to ship to production? — Extender7777 · 2026-10-28
- Lucas Meijer: writing and evaling skills forces you to say what you care about — lucasmeijer · 2026-10-28
- Merge CEO Shensi Ding stays hands-on with Codex: execs set the AI fluency ceiling — OpenAI · 2026-10-28
- Detection Isn't Blocking: How to Actually Verify Your AI Agent's Security Guard — shoumikgoswami · 2026-10-28