Executive Director Principal Software Engineer - Agentic Engineering

JPMorgan Chase & Co.; Jersey City (NJ); On-site; USD 120,000 - 160,000; Full time

Job summary

JPMorgan Chase & Co. is seeking a Principal Software Engineer in Jersey City. You will lead the design and scaling of a next-generation AI platform and be responsible for production coding, architectural decisions, and building a high-performing team. The ideal candidate has over 10 years of software engineering experience, with at least 5 years in technical leadership and expertise in Python and distributed systems. Join us to shape the future of financial technology.

Qualifications

• 10+ years in software engineering, including 5+ years leading senior technical teams.
• Deep familiarity with agentic frameworks like LangChain.
• Strong command of prompt/context design and multi-step reasoning patterns.

Responsibilities

• Own the multi-year agentic platform strategy.
• Design distributed ingestion and workflow systems.
• Contribute directly in Python and lead design/code reviews.

Skills

Expert Python

Distributed systems experience

MLOps tooling

Cloud fluency

Executive communication

Tools

Docker

Kubernetes

Kafka

Spark

Job description

If you are looking for a game-changing career, working for one of the world's leading financial institutions, you’ve come to the right place.

As the Principal Software Engineer at JPMorgan Chase within the Corporate and Investment Bank Technology – Securitized Product Group Technology team, you will lead the design, build, and scaling of our next-generationmulti-agent AI platform. This is not a manage-from-a-distance role: you will write production code, own key architectural decisions, unblock engineers, and be accountable for production outcomes—while also setting multi-year strategy and building a high-performing team.

If you’ve built agentic systems in production (not prototypes), can distinguish orchestration vs. parallelization in real distributed environments, and can whiteboard with a CTO then land a PR the same afternoon—this role is for you.

Job responsibilities

• Own the multi-year agentic platform strategy: agent toolchains, RAG pipelines, memory/state architectures, context management, evaluation, and feedback/reinforcement loops.
• Architect scalable multi-agent systems using LangChain, LangGraph, AutoGen, or equivalent frameworks—and define when to use simpler primitives.
• Design distributed ingestion and workflow systems (batch + streaming) with data contracts, lineage, and strong data-quality patterns.
• Establish standards for the agentic development lifecycle: context engineering, automated evals, observability, security, and release readiness.
• Contribute directly in Python (services, concurrency, performance, reliability) and set the bar via reference implementations.
• Lead design and code reviews; engage directly in incident response and production hardening.
• Build reusable agent components: planning/decomposition, tool/function calling, self-critique/reflection loops, state management, multi-agent coordination, and safety controls.
• Partner with MLOps/platform on deployment, monitoring, and retraining pipelines (MLflow, SageMaker, Vertex AI, Azure ML, etc.).

Required qualifications, capabilities, and skills

• 10+ years in software engineering, including 5+ years leading senior technical teams and driving architecture for large systems.
• Expert Python; working proficiency in TypeScript, Go, or Rust.
• Deep familiarity with agentic frameworks (LangChain, LangGraph, AutoGen, or equivalent).
• Strong command of RAG, prompt/context design, tool/function calling, and multi-step reasoning patterns.
• Distributed systems and data platform experience: Kafka, Spark, gRPC/REST, Docker/Kubernetes.
• Cloud fluency (AWS/Azure/GCP) and MLOps tooling (MLflow, Kubeflow, SageMaker, Vertex AI).
• Familiarity with vector databases, knowledge graphs, and semantic retrieval patterns.
• Strong executive communication—able to align stakeholders on trade-offs, timelines, and risk.
• Platform mindset: builds durable systems, not fragile demos.

Preferred qualifications, capabilities, and skills

• kdb+/q, ClickHouse, or other time-series/analytical data stores.
• Financial services domain (trading infrastructure, derivatives, fixed income).
• Publications, talks, or open-source contributions in agentic AI.
• Experience designing “agentic SDLC” / software-delivery automation.