Staff Agentic Software Engineer

Chicago Mercantile Exchange•Chicago, IL

About The Position

CME Group is seeking a senior and experienced Staff Agentic Software Engineer to join a dynamic team responsible for our mission-critical Real-time Positions & Risk Management Systems. This role is pivotal in driving both the technical evolution of our financial platforms and the transformation of our engineering workflows. You will lead the team in transitioning from traditional software development to an agentic engineering paradigm —leveraging advanced LLM coding tools (e.g., Gemini CLI, Antigravity) to build high-performance systems faster, safer, and at greater scale. As a technical leader, you will balance aggressive AI-driven velocity with the strict safety, low-latency, and zero-downtime requirements of CME Group's core clearing and risk management functions.

Requirements

Bachelor’s degree or higher in Computer Science, Mathematics, Financial Engineering, or a related field, with 8+ years of hands-on experience building, deploying, and maintaining scalable real-time systems across the full stack.
Expert-level proficiency in Java and the Spring framework.
Deep, hands-on experience designing and debugging multi-threaded concurrent applications, lock-free data structures, memory management, thread pools, and race condition diagnostics.
Hands-on experience with AI coding agents (e.g. Gemini CLI, Claude Code, Codex, etc) in real production workflows: multi-step agent tasks, agent-authored PRs, agent-driven test generation.
Deep experience with Google Cloud Platform ( GCP ) services (GKE, Pub/Sub, BigQuery, Cloud Run, Dataflow) and real-time messaging frameworks ( Kafka, MQ, Flink ).
Strong proficiency in SQL, Postgres DB, and Python (for scripting, automated evals, data analysis, or tooling integration).
Experience with distributed tracing, SLO/SLI monitoring, and chaos engineering in production environments where system failure carries direct financial or regulatory impact.
Daily operational fluency with CLI and terminal-based agent environments ( Claude Code, Gemini CLI, Codex ) as well as agentic IDEs ( Cursor, Antigravity ).
Direct experience configuring system context, building or integrating Model Context Protocol (MCP) servers, function calling, and structured domain-prompting.
Proven track record of designing property-based tests, static analysis rules, and code-review workflows specifically built to catch AI hallucinations, edge-case failures, and security vulnerabilities.
Demonstrated ability to establish team-wide AI coding norms, measure developer outcome velocity, and champion an AI-first engineering culture.

Nice To Haves

Experience developing software for financial risk management, high-frequency trading, or clearing systems.
Experience building internal developer tools, CLI extensions, or custom LLM evaluation harnesses.
Familiarity with local model deployments or fine-tuning workflows for enterprise dev environments.

Responsibilities

Lead the architecture, design, and development of high-volume, low-latency Java applications on Google Cloud Platform (GCP) for mission-critical systems, ensuring ultra-high availability, low jitter, and thread safety.
Lead the team’s shift toward agentic software engineering.
Standardize toolchains, system prompts, context repositories, and agentic loops across the development lifecycle.
Build and maintain the shared agent infrastructure: repo-level agent context, MCP server integration, codebase indexing pipelines, and local developer tooling that feed deep, domain-specific context into LLM agents.
Design dynamic evaluation harnesses, automated test suites, and CI/CD guardrails specifically tailored to audit, test, and validate AI-generated code for concurrency bugs, memory leaks, and performance regressions.
Maintain and enhance high-throughput CI/CD automation pipelines to ensure seamless, secure, and reliable software delivery into production.
Mentor engineers through the workflow shift; lead hands-on workshops to train traditional software developers into proficient AI-native engineers.
Drive architecture decisions across team boundaries and be able to articulate tradeoffs clearly to both engineers and stakeholders.
Be able to operate under pressure and on-call for system where failure has direct financial or regulatory input.
Elevate team capabilities through rigorous code reviews, design reviews, pairing, and active mentorship.
Contribute to development tooling and engineering culture initiatives.
Drive technical leadership by mentoring the team on agentic AI capabilities, accelerating feature delivery while maintaining strict code quality and reliability.