Quantitative Trading & Research - Applied Researcher – Agentic AI Systems - Associate

JPMorgan Chase & Co.•New York, NY

22d•$150,000 - $200,000

About The Position

At JPMorganChase, we’re building the next generation of AI-powered workflow automation. This is a hands-on role for someone who thrives on ambiguity, ships quickly, and is energized by hard technical challenges. As an Associate Applied Researcher in the Quantitative Trading & Research (QTR) Team, you’ll sit at the intersection of applied research and production engineering turning frontier GenAI capabilities into reliable, high-leverage agentic systems that transform how we respond to inbound client requests.

Requirements

Strong coding skills (Python preferred) and comfort owning production code
Advanced degree in Computer Science, Data Science, Machine Learning, or related field.
Hands-on experience building with LLMs (agent frameworks, tool use, RAG, prompt engineering, evals)
Strong understanding of modern GenAI capabilities, failure modes, and practical mitigation strategies
Demonstrated applied research track record delivering ML/AI systems that moved a real business or user metric
Ability to explain technical tradeoffs to non-technical stakeholders and write clearly
Bias to action; comfortable working in ambiguity with rapidly evolving requirements
Strong ownership mindset with a focus on improving what’s broken without waiting for permission

Nice To Haves

Experience with Equity Derivatives and Pricing
Familiarity with evaluation frameworks, LLM observability, or fine-tuning open-weight models
Experience scaling an agentic prototype into a production system used by real users
Experience designing and operating monitoring/QA processes for LLM outputs (quality, safety, reliability)

Responsibilities

Build agentic systems end-to-end: design, prototype, and productionize multi-step LLM agents that retrieve context and generate accurate, well-structured responses
Drive applied research by evaluating emerging techniques (tool use, planning, retrieval, evaluation frameworks, fine-tuning, prompt optimization) and integrating the best into production
Own the full loop from problem framing and dataset construction through model/agent design, evaluation, deployment, and monitoring
Improve quality systematically via evals, error analysis, and feedback loops that convert subjective issues into measurable fixes
Partner cross-functionally with sales, quant research, trading, product, and engineering to deeply understand RFQ/client workflows and ship adopted solutions
Build and maintain production-grade code and systems that are observable, robust, and scalable
Contribute to technical direction and standards for agent design, evaluation, and safe deployment