Role Summary
We are seeking a contract AI Engineer (Generative AI) to design, build, and deploy practical GenAI solutions (e.g., document Q&A/RAG, structured extraction, copilots) in collaboration with product and engineering stakeholders. This role is hands-on and delivery-oriented, with a focus on implementing reliable services, evaluation, and clear documentation for maintainability.
Key Responsibilities
- Design and implement GenAI application workflows, including RAG pipelines (ingestion, chunking, embeddings, retrieval, prompting).
- Build structured extraction solutions (documents fields/JSON) with validation logic and post-processing where required.
- Build and maintain API services (e.g., FastAPI/Flask) to expose GenAI capabilities for integration with internal systems.
- Create and maintain evaluation datasets and test harnesses (accuracy/consistency checks, regression tests, latency tracking).
- Conduct model and approach comparisons (LLM/embedding/retriever variants) and document trade-offs and recommendations.
- Apply engineering best practices: version control, reproducible environments, basic testing, logging, and error handling.
- Produce clear technical documentation: architecture notes, setup guides, runbooks, and handover materials.
- Collaborate with stakeholders to translate business needs into implementable user stories and deliver incrementally.
Required Qualifications
- 3+ years experience in a data/ML/software engineering role with strong Python development skills.
- Contract Basis - Renewable up to 2 years
- Practical experience delivering at least one of the following:
- Retrieval-Augmented Generation (RAG) using a vector database,
- LLM-based information extraction into structured outputs,
- Integration of LLMs via APIs for an end-user workflow.
- Experience building and consuming REST APIs and working with JSON schemas / structured outputs.
- Familiarity with evaluation concepts and metrics; able to implement repeatable testing for model quality.
- Comfortable working independently, communicating progress clearly, and iterating quickly based on feedback.
Preferred Qualifications
- Experience with vector stores (e.g., FAISS, Qdrant, Milvus, Pinecone) and retrieval tuning.
- Familiarity with local LLM tooling (e.g., Ollama) and/or cloud LLM platforms.
- Experience with document formats and parsing (PDF/XML/HTML), regex-based post-processing, and edge-case handling.
- Exposure to Docker, CI/CD, and basic observability/monitoring practices.
- Experience working with security/privacy requirements (PII handling, access controls).
- Experience with AWS cloud services (e.g., deploying services, using managed storage/compute, or integrating with AWS-native tooling).
Technical Skills
- Python (data handling, APIs, testing), SQL
- LLM/RAG concepts: embeddings, chunking strategies, retrieval, prompt templates
- API development (FastAPI/Flask), integration patterns
- Basic software engineering practices (Git, documentation, reproducibility)
- Strong problem-solving and debugging ability
- Clear written and verbal communication with technical and non-technical stakeholders
- Ability to manage priorities and deliver outcomes in a time-boxed environment
