Applied AI Engineer

About the role

Hybrid schedule (3 days on-site at 1633 Broadway, New York, NY) with a 12-month contract that may extend or convert to full-time employment.

Responsibilities

• Build AI or data-driven applications that integrate APIs, databases, cloud services, and enterprise data sources.
• Implement and scale Retrieval-Augmented Generation (RAG) pipelines over a large internal document repository (“Doc Central”) that currently drives ~20% of the firm’s OpenAI traffic.
• Design, test, and validate prompts; manage context and grounding; and validate outputs for reliability and safety.
• Develop agentic AI workflows using Python, OpenAI or Claude Agent SDK, and other LLM frameworks.
• Work with PostgreSQL and vector databases (Elasticsearch experience is a plus) to store, index, and retrieve embeddings at scale.
• Evaluate AI system performance and iterate on retrieval, generation, and agentic capabilities as the platform scales.

Requirements

• 5–7 years of hands-on software engineering experience (more is acceptable).
• Strong Python coding skills and core AI fundamentals.
• Direct experience building applications with GenAI and agentic AI concepts: prompt design/testing, grounding, context management, and output validation.
• Practical experience integrating LLM APIs (OpenAI, Claude, or comparable) into production applications.
• Experience with RAG pipelines and vector search over large document collections.
• Proficiency with PostgreSQL; familiarity with vector databases is acceptable.
• Excellent written and verbal communication; ability to explain technical decisions clearly.
• Curiosity and a strong understanding of why specific technical choices are made.

Preferred Qualifications

• Experience in financial services, particularly fixed income (e.g., understanding of stocks, bonds, and related markets).
• Hands-on experience with Elasticsearch and embedding workflows.
• Familiarity with developer tools such as ClaudeCode, Amp, and CodeEx.