About

We are a forward-thinking fintech startup building an intelligent financial research assistant. Our goal is to empower users with deep, contextual insights from a massive library of financial reports, market analyses, and SEC filings. We're looking for a skilled and innovative Full Stack Engineer to build the core application from the ground up. This isn't just a standard CRUD app. The heart of this project is a sophisticated AI engine that can understand complex financial queries and provide accurate, sourced answers. You will be responsible for architecting and implementing both the user-facing application and the intelligent backend that powers it. # Key Responsibilities: Frontend (Next.js 14+): Develop a beautiful, responsive, and highly interactive user interface using Next.js, React, and Tailwind CSS. Implement app router, server-side rendering (SSR) where beneficial, and clean state management. Backend & AI Integration (Python): Design and build a robust Python backend (using FastAPI or Flask) to handle business logic and, most importantly, serve as the engine for our RAG system. RAG System Development: Architect and implement the entire Retrieval-Augmented Generation pipeline. This includes: Document Processing: Ingesting and chunking PDFs and text documents. Vectorization: Creating embeddings for the text chunks. Vector Database: Storing and efficiently querying embeddings using a vector database like Pinecone or Weaviate. LLM Orchestration: Integrating with a Large Language Model (e.g., OpenAI GPT-4, Anthropic Claude) to generate final answers based on the retrieved context. End-to-End Ownership: You will be involved in the entire development lifecycle, from initial system design and database schema creation to deployment and initial troubleshooting. # What You'll Deliver (Scope of Work): This is a fixed-scope project with the following clear deliverables: A fully functional, deployed Next.js web application with a clean UI for user registration, document upload (admin), and a chat/query interface. A secure and scalable Python API backend with well-documented endpoints. A fully implemented RAG pipeline that: Accepts document uploads (primarily PDF). Processes, chunks, and creates embeddings for the documents. Stores these embeddings in a vector database. Provides a query endpoint that retrieves relevant context and generates a coherent, well-sourced answer using an LLM. A simple PostgreSQL database for storing user data and application metadata. Complete documentation and a handover session. # Skills & Qualifications: Must-Haves: Proven experience as a Full Stack Engineer with a strong portfolio. Expert-level proficiency in Next.js 14+ (App Router, SSR, API Routes). Strong Python skills and experience building APIs with FastAPI or Flask. Hands-on experience building a production-grade RAG (Retrieval-Augmented Generation) system. Practical knowledge of vector databases (e.g., Pinecone, Weaviate, or Chroma). Experience integrating and working with LLMs (OpenAI API, Anthropic, or local models). Proficiency with relational databases (PostgreSQL preferred). Ability to write clean, maintainable, and well-documented code. # Nice-to-Haves: Experience in the Fintech industry or with financial data. Knowledge of authentication and authorization best practices (NextAuth, Auth0, etc.). Experience with cloud deployment (Vercel, AWS, GCP).