Python/GenAI Engineer

What You'll Do8

• Build and ship LLM applications such as intelligent assistants, document summarizers, and workflow automation with LangChain, LlamaIndex, and hosted models from OpenAI, Anthropic, or Azure OpenAI.
• Architect retrieval-augmented generation pipelines on vector stores like Pinecone, FAISS, or pgvector, picking embedding models and chunking strategies that tie answers to enterprise data.
• Write prompts, test them, refine agentic workflows, and set up evaluation frameworks that track accuracy, relevance, hallucination rates, and latency.
• Develop Python services and RESTful APIs with FastAPI or Flask, keeping code well tested and easy to maintain.
• Create data ingestion and transformation pipelines with Pandas, SQL, and cloud data services to prepare structured and unstructured data for model workloads.
• Deploy and monitor AI systems on AWS, Azure, or GCP using Docker, CI/CD pipelines, and observability tooling that protects reliability and controls cost.
• Add guardrails, data privacy controls, and governance checks that satisfy client security, model risk management, and regulatory requirements.
• Take part in sprint planning, daily stand-ups, and Jira tracking, and translate business needs from client and cross-functional partners into working AI solutions.

Requirements8

• Hands-on Python experience building production services, not just notebooks or scripts.
• Track record with LLM application frameworks such as LangChain or LlamaIndex and with commercial model APIs from OpenAI, Anthropic, or Azure OpenAI.
• Practical knowledge of retrieval-augmented generation, covering vector databases, embedding models, and chunking approaches.
• Comfort writing and tuning prompts, plus setting up evaluations that measure hallucination rates and latency alongside answer quality.
• API development with FastAPI or Flask and solid testing habits.
• Data work with Pandas and SQL across structured and unstructured sources.
• Cloud deployment on AWS, Azure, or GCP with Docker and CI/CD pipelines.
• Client-facing communication skills for financial services stakeholders, plus familiarity with security, privacy, and model risk controls.

Who Should Apply

Engineers who can carry an LLM proof of concept through to a governed production service suit this role best. You bring Python backend depth, RAG pipeline experience, and the confidence to work directly with financial services clients. Freelancers near New York or Jersey City who accept hybrid, client-site work should apply.

Salary Insight