Hailo Training Pipeline ↗
A model preparation and optimisation pipeline for Hailo AI accelerators.
Personal repository / AI & vision
Rotoclear AI Chatbot
A retrieval-augmented assistant connecting product knowledge with natural-language questions.
Discuss the work ↗SYSTEM MAPSelect a component to explore ↙
From source to answer
Web pages
Cheerio extracts source material for ingestion.
Simplified architecture · illustrative, not a product screenshot
01 / THE CONTEXTA closer look at
A closer look at
the work.
A retrieval-augmented assistant connecting product knowledge with natural-language questions.
TOOLS & TECHNOLOGIES
SvelteKitLangChainSupabaseRAG
A personal repository in my connected GitHub account.
02 / INSIDE THE PROJECT
Behind the diagram.
Technical notes, when you want to go deeper.
01Implementation
Product knowledge is ingested from web pages, split into passages and indexed as embeddings in Supabase pgvector. A SvelteKit interface connects questions to a LangChain retrieval flow and an OpenAI model.
02Key decision
Retrieval gives the answer-generation step selected context instead of relying only on a model’s general knowledge. Content ingestion remains a separate responsibility: repeat imports need curation to avoid duplicate material.
03Current scope
The repository documents a retrieval-augmented assistant. Answer accuracy, source freshness and real-world support outcomes have not been independently measured for this case study.
REPOSITORY NOTES
A closer look.
Source code is private. The portfolio contains a high-level description, without internal files or customer data.
Discuss the work ↗- Repository
- rotoclear-chatbot
- Editorial basis
- Repository documentation
- Default branch
- master