Campus RAG Assistant

Production-style enterprise RAG platform for governed campus knowledge.

Campus RAG Assistant demonstrates how institutional knowledge can be served through a measurable, observable, citation-first RAG platform. It combines a Vue product UI, FastAPI backend, AWS / Azure / mock provider boundaries, LangGraph orchestration, RAGAS evaluation, LangSmith traces, CI/CD, load testing, and operational hardening docs.

Portfolio focus: Lead AI Engineering and AI Platform Architecture.

Review model

This project is evaluated through source code, architecture docs, screenshots, evaluation results, and operational artifacts. It is not presented as a hosted public product.

View source on GitHub ->

Start here

If you are...	Read
Short on time	Reviewer Guide
Evaluating ownership and judgment	Case Study
Reviewing architecture	Architecture + Design Notes
Reviewing AI quality	Evaluation + Baseline
Reviewing production maturity	CI/CD + Operations + Security

Senior engineering signals

Area	Signal
Platform architecture	Provider registry isolates AWS, Azure, and mock execution modes
RAG orchestration	LangGraph exposes retrieval stages for tuning and traceability
Evaluation	RAGAS baseline is documented honestly, with gates used as release controls
Observability	LangSmith spans, Prometheus metrics, request IDs, structured logs
Product judgment	KB-first answers, cited sources, opt-in web research, visible disclaimer
CI safety	Mock providers allow tests without cloud credentials

What this shows

Layer	What is demonstrated
Product	Governed KB-first chat, cited sources, opt-in web research, feedback loop, and campus-ready UX
RAG engineering	LangGraph retrieval stages, multi-query retrieval, rerank hooks, fallback chain streaming, and explicit source contracts
Platform architecture	AWS/Azure/mock provider registry, tenant config, feature flags, Alembic migrations, and CI-safe local mode
Evaluation	RAGAS golden-set regression harness, documented Phase 5 baseline, and LangSmith traces for KB/web paths
Operations	GitHub Actions, gitleaks, dependency review, Prometheus metrics, k6 load tests, release docs, and runbooks

Quality baseline

The project includes a RAGAS golden-set harness and a documented baseline. Phase 5 retrieval tuning improved AWS context recall to 0.800, passing the retrieval coverage gate. Context precision remains the main improvement target; next work focuses on ingestion/chunking and rerank tuning.

This is intentionally presented as an engineering baseline, not a marketing claim. Strict RAGAS gates are release controls, not blockers for local demo or ordinary PR CI.

Read more: Evaluation approach and baseline scores.

Architecture

flowchart LR
  VueSPA["Vue 3 SPA"] --> FastAPI["FastAPI"]
  FastAPI --> LangGraph["LangGraph RAG"]
  LangGraph --> Providers["Provider Registry"]
  Providers --> BedrockKB["AWS Bedrock KB"]
  Providers --> AzureAI["Azure AI Search"]
  FastAPI --> Postgres[("Postgres")]
  FastAPI --> Prometheus["Prometheus"]
  LangGraph --> LangSmith["LangSmith"]
  LangGraph --> RAGAS["RAGAS eval"]

Design detail: Architecture and Design Notes.

Documentation

Visitor	Best entry point
New here	This page
Hiring / portfolio reviewer	Case Study
Architecture reviewer	Architecture, Design Notes, ADRs
Evaluation reviewer	Evaluation Approach, Evaluation Baseline
Operations reviewer	Operations, CI/CD, Release, Security
Product demo reviewer	Screenshots and demo script
Roadmap reviewer	Product Roadmap

Screenshots

Knowledge-base answer

Source transparency

Opt-in web research

LangSmith trace

More assets: screenshots catalog.

Stack

Layer	Technologies
Backend	FastAPI, SQLAlchemy, Alembic, JWT auth, rate limiting, Prometheus metrics
Frontend	Vue 3, TypeScript, Pinia, Tailwind, Vitest, Playwright
RAG orchestration	LangGraph (RAG_ENGINE=langgraph) or LangChain ConversationalRetrievalChain (RAG_ENGINE=chain)
Retrieval	Bedrock KB / OpenSearch Serverless, Azure AI Search, multi-query + RRF, optional rerank
LLM	AWS Bedrock, Azure OpenAI, or mock provider
Web search	Mock or Tavily behind research_mode=web
Eval	RAGAS golden dataset, tox -e eval, LangSmith traces
CI/CD	GitHub Actions, tox, gitleaks, dependency review, optional EB deploy

Feature availability

Configuration	What works
No cloud keys (RAG_FORCE_MOCK=true)	Register/login, chat UX, streaming path, source panel, feedback, local tests
AWS Bedrock KB	Managed KB retrieval, Bedrock generation, LangGraph retrieval stages, LangSmith trace capture
Azure OpenAI + AI Search	Azure provider path with vector/keyword/hybrid retrieval and cited answers
Web research enabled	Per-message web mode with disclaimer UI and WEB-labeled sources (mock or Tavily)
OAuth configured	GitHub OAuth handoff to Vue; Google-ready provider config
Eval keys available	RAGAS golden-set runs, release quality gates, LangSmith trace inspection

Getting started

python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
# set RAG_FORCE_MOCK=true, LLM_PROVIDER=mock, RETRIEVER_PROVIDER=mock
createdb chatbot_dev
alembic upgrade head
./scripts/run-backend-venv.sh          # http://127.0.0.1:8000
./scripts/run-frontend-vue.sh          # http://127.0.0.1:5173

Register a user and start a chat. Responses use the mock provider unless you configure live AWS/Azure providers.

Review artifacts

Artifact	Status
Source code	Implemented
Vue product UI screenshots	Included
Local mock execution path	Implemented
AWS Bedrock KB path	Implemented
Azure AI Search / OpenAI path	Implemented
RAGAS baseline	Documented
LangSmith traces	Captured in screenshots
Public hosted product	Not claimed
Official campus deployment	Not claimed

Origin and Scope

This repository builds from the public ets-berkeley-edu/chabot codebase and substantially extends it as an independent portfolio and educational project. The work here focuses on the AI platform surface: Vue product UI, provider abstraction, LangGraph orchestration, RAGAS evaluation, LangSmith observability, CI/CD, load testing, and operational documentation. It is not an official UC Berkeley or UC product.

See Notice for attribution details.

License

Software in this repository is licensed under the Regents of the University of California terms. See Notice for attribution details. Commercial use requires an agreement with UC OTL.