Applied AI Lead Engineer - Agentic Systems

JPMorgan Chase JPMorgan Chase · Banking · LONDON, LONDON, United Kingdom · Commercial & Investment Bank

Lead Engineer role focused on designing, building, and productionizing autonomous and assistive AI agents and multi-agent systems. This involves implementing RAG, integrating tools, prompt engineering, guardrails, and orchestrating workflows, with a strong emphasis on scaling and reliability within a fintech domain.

What you'd actually do

  1. Architect, develop, and productionize autonomous and assistive AI agents to enhance operations.
  2. Design multi-agent systems, including role definition, tool integration, planning, memory, and workflow orchestration using modern agent frameworks.
  3. Implement Retrieval-Augmented Generation (RAG) pipelines and semantic search with vector databases, including indexing, retrieval policies, and evaluation.
  4. Build and integrate agent tools and APIs to connect agents with external services, databases, and internal systems, ensuring robust output parsing and error handling.
  5. Practice advanced prompt engineering and implement output validation and guardrails to reduce hallucinations.

Skills

Required

  • Experience building and deploying agentic AI applications in production environments.
  • Expertise with ML frameworks such as PyTorch, TensorFlow, and scikit-learn.
  • Proficiency in Python; experience writing comprehensive tests and building evaluation harnesses for agents and prompts.
  • Hands-on experience with agent frameworks such as LangChain, CrewAI, AutoGen, LangGraph.
  • Knowledge of generative models including transformers, GANs, VAEs, and diffusion models.
  • Understanding of data preprocessing, feature engineering, and model/agent evaluation techniques.
  • Familiarity with cloud platforms and containerization technologies.
  • Strong problem-solving skills and ability to work independently and collaboratively.
  • Effective communication skills for technical and non-technical audiences.

Nice to have

  • Experience in financial services, especially investment banking operations.
  • Experience fine-tuning small language models with approaches like LoRA, QLoRA, DoRA; quantization and distillation.
  • Familiarity with prompt optimization frameworks and building prompt pipelines and evaluation suites.
  • Experience with distributed computing, data sharding, and performance optimization.
  • Hands-on experience with AWS services related to AI deployment and workflow orchestration.

What the JD emphasized

  • production environments
  • agent frameworks
  • evaluation harnesses for agents and prompts
  • agentic AI applications

Other signals

  • productionize autonomous and assistive AI agents
  • multi-agent systems
  • agent frameworks
  • RAG pipelines
  • vector databases
  • agent tools and APIs
  • prompt engineering
  • output validation and guardrails
  • multi-step workflows
  • human-AI interfaces
  • performance evaluations