AI Engineer

Job Description

AI Engineer (GenAI Workflows for OTC Derivatives)

\n

Location: New York City (SoHo) — In-Person

\n

\n

About Our Client

\n

Our client is building the next-generation platform for institutional finance. The OTC derivatives market moves trillions of dollars daily on infrastructure that hasn’t fundamentally changed in decades—and this team is rebuilding that foundation with a modern stack , a delightful product experience , and an AI-native architecture .

\n

\n

The founding team comes from trading, quantitative, and engineering roles at tier-one financial institutions and has built some of the most innovative products in the market. Their clients include sophisticated banks, hedge funds, and asset managers—meaning the problems are complex, the scale is massive, and the software will shape how this market operates for years to come.

\n

\n

Tech Stack

\n

\n
• Cloud-native, serverless-first architecture + Infrastructure-as-Code (IaC)
\n
• Backend services in Python and Java
\n
• Frontend in TypeScript / React
\n
• Modern foundation models + AI tooling
\n
• Automated SDLC using tools like GitHub , Logfire , Vercel , and rapid intraday deployment practices
\n

\n

\n

The Role

\n

\n

Our client is hiring an AI Engineer to build and productionize generative AI workflows that increase efficiency across the OTC derivatives lifecycle. You’ll own key parts of the evaluation and context engineering infrastructure—ensuring model outputs are high-quality, reliable, and explainable.

\n

\n

You’ll work closely with financial engineers and domain experts to design retrieval, grounding, and validation pipelines for complex derivatives data. The work includes building tools for knowledge extraction from trade documents, improving data consistency across the trade lifecycle, and integrating AI reasoning into workflows like settlement, valuation, and risk management.

\n

\n

What You’ll Do

\n

\n
• Build GenAI workflows that automate and enhance OTC derivatives processes end-to-end
\n
• Own context engineering (retrieval, grounding, memory, and data shaping) to improve response quality and reliability
\n
• Build evaluation systems (offline + online) to measure accuracy, consistency, and explainability
\n
• Design structured output patterns (schemas, constrained generation, validation) for production-grade AI
\n
• Create pipelines to extract and normalize data from unstructured trade documents (confirmations, term sheets, etc.)
\n
• Integrate AI reasoning into downstream workflows (settlement, valuation, risk, cashflows, reconciliation)
\n
• Collaborate with domain experts to translate financial requirements into robust AI systems
\n
• Contribute to a high-velocity engineering culture with strong SDLC practices and rapid deployments
\n

\n

\n

Requirements

\n

\n
• Bachelor’s degree in Computer Science, Engineering, Mathematics, Financial Engineering, or related field
\n
• In-person role in NYC (SoHo)
\n
• Strong proficiency in Python (Java experience is a plus)
\n
• Proficiency in prompt engineering and structured output generation (JSON, Pydantic schemas, validation)
\n
• Experience with NoSQL databases (e.g., MongoDB , DynamoDB )
\n
• Familiarity with cloud infrastructure and deployment tools (CDK, Terraform) and cloud providers (AWS, GCP, Azure)
\n
• Strong understanding of both traditional SDLC and AI SDLC (versioning, testing, rollout, monitoring)
\n
• Strong software design fundamentals
\n
• Excellent problem-solving and communication skills
\n

\n

\n

Bonus Qualifications

\n

\n
• Experience fine-tuning or deploying LLMs (OpenAI, Anthropic, Gemini, etc.)
\n
• Familiarity with vector databases and RAG pipelines
\n
• Experience building AI agents or workflow orchestration for financial data tasks
\n
• Understanding of OTC derivatives data models and trade representations (e.g., ISDA concepts)
\n
• Experience integrating AI systems with pricing, settlements, risk, or cashflow infrastructure
\n
• Familiarity with LangChain, PydanticAI, LlamaIndex, or similar frameworks
\n
• Experience extracting data from unstructured financial documents (confirmations, term sheets, schedules)
\n
• Strong grasp of evaluation, hallucination mitigation, and data quality management in production AI systems
\n

\n

\n

Why This Role

\n

This is a rare chance to build production-grade AI systems in a domain where correctness matters and the upside is massive. If you want to own foundational AI infrastructure, work alongside domain experts, and modernize a trillion-dollar market—this role is built for high-impact engineers.