ProjectIntent Over Keywords: From Stitched Embeddings to a Production Fashion Agent · Module 17: Grounded generation · part 19 of 26
Part 19 · Module 17: Grounded generation

Module 17: Grounded generation

26 min read·
About 50 minutes · Needs: Module 9 (grounded answers in Part 2), Module 14 (the planner), Module 16 (product search) · Code: part3-styleme/backend/src/styleme/agent/generator.py, agent/grounding.py, agent/parsing.py, agent/prompts.py

Where this fits

WHERE THIS FITS Intent Over Keywords this module PART 1: STITCHED VIDEO RAG three models, two spaces, RRF PART 2: UNIFIED VIDEO RAG one model, one space PART 3: STYLEME FastAPI, LangGraph, React QUALITY AND OPERATIONS evals, Docker, Cloud Run Frames ffmpeg, one per 5 s CLIP ViT-B/32 frames + text, 512 d Whisper base speech to text MiniLM 64-word chunks, 384 d Two collections p1_visual, p1_text Rank fusion RRF, k = 60 Gemini answer from transcript text API + Streamlit upload, ask, sources Windows ffmpeg, video + audio Embedding 2 Gemini, one vector One collection p2_unified, no fusion Notes + answer Gemini, cited times React web chat, photo, chips nginx one URL, /api proxy FastAPI /api/chat, limits Sessions in memory, cap of 3 Planner intent before retrieval Implementer batched embed, search Generator stylist reply, ids Grounding check no LLM, drops invented Gemini flash-lite, embedding 2 Qdrant styleme_products Catalogue ingest image + title vectors Langfuse optional traces Test video 26 questions, 4 kinds Video eval Part 1 vs Part 2 Agent evals intent, grounding Guards uploads, limits, secrets Containers compose profiles Cloud Run StyleMe only CI/CD GitHub Actions, WIF Original code v1, v2, fashion-agent
This module (highlighted): the generator, the Gemini model it calls (amber, a model) and the grounding check (green, a safeguard) that reads every reply before the shopper does.

This is the last step of a search turn. The planner has decided to search (Module 14), the implementer has returned up to 8 products (Module 16), and the generator writes the reply the shopper reads. The highlighted parts are the generator, the Gemini model it calls, and the grounding check that reads the model's reply before anyone else does.

The rest of this project is yours to keep

This project is bought on its own, once, and stays readable afterwards, including the parts added to it later.

Or open everything with a plan