About the job
We have an exciting and rewarding opportunity for you to take your data engineering career to the next level. As a Lead Software Engineer - Databricks/PySpark/AI at JPMorganChase within the Corporate Sector-Global Finance team, you will serve as a senior hands-on developer and technical leader within an agile team, responsible for building, delivering, and optimizing cutting-edge data products that power agentic AI systems — autonomous AI agents capable of planning, reasoning, and executing multi-step tasks. In this role, you will write production-quality code daily, drive implementation of essential technology solutions including data infrastructure, tool integrations, and retrieval systems that enable AI agents to access, interpret, and act on enterprise data in support of the firm’s business goals. You will be expected to mentor junior engineers, collaborate with cross-functional stakeholders, and champion engineering excellence through hands-on delivery.
Responsibilities
Building and optimizing data pipelines and workflows that serve as the backbone for agentic AI systems, ensuring agents have reliable, real-time access to high-quality, structured and unstructured data
Developing data retrieval and indexing layers that enable AI agents to autonomously search, query, and synthesize information across multiple data sources
Building and maintaining tool-use infrastructure — APIs, data services, and function endpoints — that AI agents invoke to execute tasks, retrieve data, and interact with enterprise systems
Implementing and enforcing best practices for data management, ensuring data quality, security, and compliance, including governance of data consumed and generated by autonomous AI agents
Hands-on development of secure, high-quality production code following AWS best practices, and deploying efficiently using CI/CD pipelines;
Building orchestration and state management layers that support multi-step agent workflows, including memory, context persistence, and task chaining
Writing and reviewing code daily, conducting thorough code reviews, and raising the technical bar across the team;
Mentoring and guiding junior and mid-level engineers through pairing, code reviews, and technical coaching
Collaborating with product owners, data scientists, and business stakeholders to translate business requirements into working, production-ready agentic AI solutions;
Evaluating and adopting emerging agentic AI frameworks, tools, and data engineering practices to continuously improve the team’s development capabilities
Qualifications
Minimum
Formal training or certification on software engineering concepts and 5+ years applied experience
Expert-level programming skills in Python/PySpark with a strong portfolio of production-grade code
Extensive hands-on experience with Databricks and the AWS cloud ecosystem, including AWS Glue, S3, SQS/SNS, Lambda
Deep expertise with Spark and SQL
Strong hands-on experience with Lakehouse/Delta Lake architecture, application development, testing, and ensuring operational stability; Snowflake, Terraform and LLMs; Data Observability, Data Quality, Query Optimization & Cost Optimization
In-depth knowledge of Big Data and data warehousing concepts at enterprise scale
Extensive experience with CI/CD processes and automated testing frameworks
Solid understanding of agile methodologies, including DevOps practices, application resiliency, and security measures
Understanding of agentic AI concepts — how autonomous AI agents plan, reason, use tools, and execute multi-step workflows — and the data infrastructure required to support them
Experience building APIs, data services, and retrieval systems that serve as the connective tissue between AI agents and enterprise data
Demonstrated ability to lead by example through code, mentor engineers, and drive delivery across the team
Preferred
Experience with agentic AI frameworks (e.g., LangGraph, AutoGen, CrewAI, OpenAI Assistants API) and understanding of how data engineering underpins agent orchestration
Familiarity with tool-use and function-calling patterns for LLM-based agents, including building and exposing APIs and data endpoints that agents can invoke autonomously
Experience with vector databases (e.g., Pinecone, FAISS, Chroma) and embedding workflows for powering agent memory, semantic search, and retrieval-augmented generation (RAG)
Exposure to agent memory and state management patterns — short-term context windows, long-term persistent memory stores, and conversation/task history management
Familiarity with guardrails and safety frameworks for autonomous AI systems, including input/output validation, action approval workflows, and human-in-the-loop controls
Understanding of observability and monitoring for agentic systems — tracing agent decision paths, logging tool invocations, and debugging multi-step autonomous workflows
Understanding of responsible AI principles, particularly around autonomous decision-making, data provenance, and auditability of agent actions