Answer support chats from your docs and escalate to a human with full context
Intermediate18h1–3
RAGLLMVector DB
16/20New Hire Onboarding Buddy
A chat buddy answers newcomer questions from your docs and checks off onboarding tasks.
Intermediate16h1–3 people17/20 fit
Build New Hire Onboarding Buddy for startups and student clubs in ~16h with 1–3 people. Stack: RAG + Slack API + LLM + Vector DB.
Why it wins: Practical for every team, including the judges' own.
Build flow
01 · Problem
New members ask the same questions and seniors lose hours re-explaining.
02 · Solution
A RAG agent answers from handbook docs, tracks a personal onboarding checklist and nudges people about pending steps in chat.
03 · Build
- 1Index the handbook
- 2Build a Slack bot
- 3Track checklist state
- 4Report unknown questions
04 · Demo
Ask 'how do I get repo access?' and see the checklist update.
MVP checklist
- Docs-grounded answers
- Personal checklist
- Nudges
- Unanswered question report
Tech stack
| Tech | Role | |
|---|---|---|
| RAG | Retrieval over your docs | AI |
| Slack API | Chat integration | API |
| LLM | Language model | AI |
| Vector DB | Embedding search | Data |
Reference repos
Open-source projects you can build on. Links checked 2026-10-31.
Related ideas
More AI AgentsAI Agents
NGO Grant Application Agent
The agent drafts grant applications from your past reports and the funder's criteria
Intermediate18h1–3RAGLLMMarkItDown18/20AI Agents
Research Paper Reading Agent
The agent walks you through a paper section by section and builds a glossary
Intermediate16h1–2MarkItDownRAGLLM17/20