We are looking for an experienced developer to build an MVP of an internal AI-powered tool for hotel asset management. This is not a chatbot project. The goal is to build a structured application that: - analyzes hotel performance data (KPIs vs budget/forecast) - reads and searches documents (monthly reports, contracts) - generates owner-side analysis and challenge memos Scope of work (MVP): You will build a simple web-based tool that: 1) Data ingestion Upload PDFs (monthly reports, contracts) Upload Excel files (KPIs, budgets, forecasts) Uploads PPT files (commentary) 2) Data storage PostgreSQL database (structured KPI data) Document storage + parsing Vector search using embeddings (pgvector preferred) 3) AI integration Use OpenAI API (Responses API + embeddings) Implement: KPI analysis document retrieval memo generation 4) Core features Select hotel + reporting period Click “Generate Memo” Output structured analysis: Headline Key findings Evidence-backed insights Recommendations 5) Basic UI Upload page Analysis page Output page (clean, readable, with references) Tech stack (Preferred): Python (FastAPI) PostgreSQL + pgvector React (or simple frontend) OpenAI API Cloud storage (e.g. S3) What we care about: Clean, simple architecture Strong handling of data + document retrieval Reliable AI outputs (no hallucinations) Practical, usable MVP (not overengineered) Nice to have: Experience with AI agents or LLM apps Experience with document parsing / embeddings Prior work with financial or analytics tools To apply, please include: Examples of relevant projects (AI + data apps) Your experience with OpenAI or LLM APIs How you would approach document search + retrieval Estimated timeline and cost Important: We are building a practical internal tool, not a research experiment. Keep it simple, focused, and usable. Looking forward to your proposals.