AI Engineer - Python / LLM / RAG

We're looking for a hands-on AI Engineer to join Docuverse, an AI-powered document intelligence platform built for consulting, nonprofit, and international development firms. We're looking for someone who enjoys building practical AI applications and production features, rather than working on model training or AI research.

We're particularly interested in engineers who can walk us through a real LLM application they have built, explain how their RAG pipeline works, and demonstrate hands-on experience dealing with messy real-world documents. This role is focused on building AI applications on top of existing LLM services. It does not involve model training, fine-tuning, or ML research.

Responsibilities:

• Develop and enhance backend features using Python, FastAPI, or Flask.
• Build and improve RAG/retrieval pipelines that work with external client storage systems through APIs.
• Process PDF and DOCX documents, including RFPs, TORs, proposals, CVs, project sheets, tables, and scanned content.
• Work with LLM APIs such as Anthropic Claude, OpenAI GPT, and Google Gemini.
• Design prompts, structured JSON outputs, and section-wise generation workflows.
• Build AI workflows that provide source-cited outputs.
• Develop opportunity-screening workflows using public tender/funding databases, email, and other data sources.
• Extract, normalise, match, score, and recommend opportunities using LLM and rule-based workflows.
• Integrate third-party APIs and external data sources.
• Support live product pilots, troubleshoot issues, improve output quality, and ship product improvements.
• Write clean, maintainable, well-documented code and maintain it on GitHub.
• Follow strict client data confidentiality and NDA requirements.

Requirements:

• 3+ years of hands-on software development experience.
• Strong Python development skills with experience in FastAPI, Flask, REST APIs, JSON, and Webhooks.
• Hands-on experience building applications using LLM APIs, including OpenAI, Anthropic Claude, and Google Gemini.
• Practical experience with RAG, Retrieval, Prompt Engineering, and Function Calling.
• Experience working with real-world PDF and Word documents, including OCR and document processing.
• Strong understanding of third-party integrations and OAuth 2.0.
• Experience with Git and GitHub.
• Comfortable working independently in a fast-moving product environment.
• Strong problem-solving and ownership mindset.
• Clear written and spoken English.

Good to Have:

• Experience with Microsoft Graph, SharePoint, OneDrive, Google Drive, Gmail, and Outlook APIs.
• Proficiency in vector databases, including pgvector, Pinecone, Qdrant, Weaviate, and Chroma.
• Knowledge of embeddings, chunking, hybrid search, and reranking.
• Experience with LangChain and LlamaIndex.
• Familiarity with Docker and AWS, Azure, and GCP.
• Basic knowledge of React and TypeScript.
• Familiarity with Claude Code and Cursor.
• Experience with RFPs, proposals, tenders, or international development procurement.