We are developing an AI-powered knowledge platform that integrates data from various sources such as code repositories, documentation, and project management tools including Git, Jira, Azure DevOps, Quickbase, Office documents, and meeting transcripts. This platform enables AI to provide accurate, source-based answers for requirements gathering, codebase analysis, and AI-assisted software development. As a key contributor, you will take full ownership of the platform, managing everything from the ingestion pipeline and multi-source retrieval-augmented generation (RAG) workflows to the full-stack application layer.
Key Responsibilities
- Develop a unified, incremental ingestion pipeline utilizing APIs, webhooks, and connectors. This includes tasks such as parsing, normalization, deduplication, and metadata extraction to ensure clean and structured data flow.
- Design and implement multi-source RAG workflows involving chunking, embeddings, indexing, hybrid retrieval, reranking, and context assembly to optimize AI-driven search and retrieval.
- Enable AI comprehension of large codebases by analyzing structure, symbols, dependencies, and APIs, and automate the generation of architecture documentation.
- Build backend services using Python (FastAPI) and frontend applications with JavaScript/Node.js to support search, data intake, and visualization functionalities.
- Ensure permission-aware data retrieval, provenance tracking, and auditability throughout the entire data pipeline to maintain security and compliance.
- Apply test-driven development (TDD), automated testing—including retrieval-quality tests—and continuous integration/continuous deployment (CI/CD) practices using Azure DevOps.
Required Qualifications
- Minimum of 5 years of strong experience in Python and backend development.
- Proven hands-on experience in building large language model (LLM) applications and RAG pipelines.
- Expertise in data ingestion, RESTful APIs, webhooks, and integrating third-party services.
- Full-stack development skills, particularly with JavaScript and Node.js.
- Familiarity with Azure DevOps, Git workflows, and CI/CD pipelines.
- Solid understanding of authentication, authorization, and secure data access mechanisms.
Preferred Qualifications and Benefits
- Experience with FastAPI, Pydantic, LangChain, or LlamaIndex frameworks.
- Knowledge of vector databases such as Pinecone, ChromaDB, or FAISS, along with retrieval evaluation techniques.
- Background in abstract syntax tree (AST) parsing, symbol indexing, dependency graphs, and knowledge graphs.
- Proficiency with Docker, cloud infrastructure, and observability tools.
- Familiarity with Jira and Quickbase integrations, as well as AI governance controls.
Joining this team offers the opportunity to build a high-impact AI platform from the ground up. You will collaborate closely with architects and product stakeholders in an iterative, security-first environment, contributing to cutting-edge technology that transforms software development workflows.