Sr Lead Software Engineer - Agentic Pricing & AI/ML

JPMorganChase•Jersey City, NJ

About The Position

As a Senior Lead Software Engineer within JPMorganChase’s Commercial & Investment Bank, Payments Technology, you will help create a standard platform for building and running AI agents in production. The platform provides a shared runtime and capabilities that support taking agents from idea to live, trusted services. As a senior technical authority for the Jersey City team, you will help shape the architecture for agentic pricing and for training, serving, and governing pricing and optimization models used as tools by AI agents. Your technical decisions will support agent-assisted pricing while maintaining the accuracy, explainability, security, and auditability required of pricing systems.

Requirements

Formal training or certification in software engineering concepts and 10 or more years of applied experience.
Hands-on experience delivering system design, application development, testing, technical troubleshooting, and operational stability.
Advanced proficiency in one or more programming languages, including Python.
Experience architecting and delivering production LLM agents or multi-agent systems, including tool use, memory, evaluation, and guardrails.
Experience designing end-to-end machine learning platforms or MLOps pipelines spanning model training, deployment, serving, monitoring, and governance.
Experience architecting systems requiring deterministic, explainable, and auditable outputs, such as pricing, billing, valuation, or risk systems.
Experience designing secure distributed systems, including identity, authorization, service-to-service trust, and controls for sensitive data.
Demonstrated experience using AI-assisted software development tools and establishing validation of AI-generated outputs for correctness, performance, security, and resiliency.
Proficiency across the Software Development Life Cycle, including agile methodologies, CI/CD, application resiliency, security, and operational practices.
Practical cloud-native experience, including Kubernetes and Amazon EKS, with knowledge of workload identity, network security, fine-grained authorization, runtime isolation, and application-level controls.
Knowledge of financial services technology environments and experience designing secure, resilient systems that support complex business processes.

Nice To Haves

Strong understanding of concurrency and distributed computation, including multithreading, asynchronous models, process and worker architectures, task queues, backpressure, and failure modes such as races, deadlocks, retries, and idempotency.
Experience with agent protocols and frameworks such as MCP, A2A, AG-UI, Google ADK, or LangGraph.
Experience with regulated workflows in which LLM outputs must be constrained, attributable, explainable, and auditable.
Experience with authorization, service-mesh, or workload-isolation technologies such as OpenFGA, OPA/Rego, Istio/Envoy mTLS, Kata Containers, gVisor, or Firecracker.
Experience with Databricks, MLflow, Kubernetes-based model serving, GPU capacity planning, or small language model inference.
Experience with optimization, pricing, revenue management, elasticity or econometric models, knowledge graphs, GraphRAG, or organizational memory in a production environment.
Experience with model risk management or comparable governance processes for models influencing revenue or client pricing, and/or knowledge of commercial card, treasury services, merchant pricing, billing, or revenue systems.

Responsibilities

Set technical direction and reference architecture for agentic pricing, including multi-agent topology, agent-to-agent communication, Model Context Protocol tool integration, and boundaries between LLM reasoning and deterministic pricing engines.
Design controls for pricing integrity, including versioned and effective-dated pricing logic, audit trails, price explanations, and separation between proposing and approving prices.
Design secure authorization and information barriers for agents operating across clients, sellers, and regions, including fine-grained authorization and isolated execution of sensitive pricing data.
Define MLOps architecture for models used as agent tools, including training, environment promotion, model registry, Kubernetes-based serving, monitoring, challenger models, back-testing, and retraining.
Establish engineering standards for agent quality and safety, including evaluation frameworks, guardrails, observability, and evidence for model risk review.
Develop creative software solutions and provide design, development, and technical troubleshooting for complex technical problems.
Develop secure, high-quality production code for critical systems; review, troubleshoot, and debug code developed by others.
Partner with platform, engineering, pricing, product, finance, and risk teams to develop reusable capabilities, shared agent and MLOps patterns, and effective governance of pricing decisions.
Drive adoption of enterprise-authorized AI-assisted engineering practices and establish validation standards for secure coding, peer review, automated testing, correctness, performance, and security.
Apply expertise across the Software Development Life Cycle and identify opportunities to automate or eliminate recurring issues to improve application and system stability.
Provide technical leadership through architecture and technology evaluations, design reviews, documented architecture decisions, and mentoring engineers.