AskRADS: An AI Recommendation Agent for Maximizing ROI of Data Science Collaborations
summary
An AI recommendation agent named AskRADS recommends experts and prior solutions from a 500-member RADS library to cut redundant data‑science work and speed decisions, claiming measurable ROI by routing requests and producing actionable problem formulations. Architecture uses R Shiny + Python, FastAPI, LangGraph with Neo4j GraphRAG and MySQL/Postgres caching, combines retrieval‑augmented generation, agent orchestration, and human‑in‑the‑loop validation, and reports successful reductions in reinventing solutions across the community.
AI-written summary · about
chapters (7)
- 0:00AskRADS introduction
- 2:04Impactful AI use cases
- 7:13Vision for collaboration
- 10:29Building AI platform
- 13:46AI agents for unstructured data
- 16:35Train agent not model
- 18:03Q&A
related talks
23mTrusted Agentic AI:Architecting Systems with Reliable Data at Scale
30mRyoma: Building Open Source AI Agents for Enterprise Data Analysis
45mUsing Postgres + OpenAI to power your AI Recommendation...
25mHow AI can help saving time, money, and your customers information?
32mRag Demystified
28mBeyond Dashboards: Architecting AI Agents for Autonomous Observability
01h 29mBuild Your Own Data Science AI Agents
29mThe GraphRAG System That Asks Back: Intelligent Probing with Contextual Answers