Agentic AI Engineering Lead — Enterprise Agent Orchestration - SVP
Citi · United States · 12d ago
The Department
Developer Engineering is a function of the CTO organization. Our mission is to make it easy and enjoyable for software engineering teams to go from a business idea to delivering an innovative product solution. We are committed to modernizing our toolchain, streamlining delivery processes, automating at scale, and embedding intelligent controls that help engineering teams ship with confidence and speed.
The Team
Within the Developer Engineering department, the Developer Services group is a dedicated expert team at the forefront of the everything-as-code agenda. We exist to deliver measurable reductions in process friction, manual effort, and human error — ensuring our policies, standards, and controls are codified, automated, and consistently applied across the organization.
We hold a unique mandate as part of a greenfield program to shape critical technical consensus at global scale, transforming how the firm's engineers build and apply controls across its technology landscape.
Developer Services is a cross-functional team of engineers, AI practitioners, data scientists, business analysts, and product managers — working together to engineer next-generation codified controls, build the platforms that power them, and drive enterprise adoption.
The Opportunity
This is a rare greenfield / build-from-scratch opportunity to create an enterprise agentic AI capability — the agent orchestration layer, the guardrails, and the adoption model — inside one of the world's most highly regulated financial environments.
As the Agentic AI Engineering Lead, you will own the enterprise rollout of autonomous and semi-autonomous AI agents across the software delivery lifecycle: designing the multi-agent orchestration platform that plans, routes, executes, and audits agent work, and defining the control model that makes AI autonomy safe under financial services regulation. You will start from a blank page, set the technical strategy, lead a team of engineers, and partner with product, security, risk, controls, and platform teams to take agents from pilot to firm-wide production.
You will work hands-on with frontier Generative AI technologies — large language models (LLMs) such as GPT, Gemini, and Claude — through prompt engineering, context engineering, tool / function calling, retrieval-augmented generation (RAG), agentic planning loops, and evaluation harnesses that prove agent behaviour before production.
This is not just an engineering role. It is a leadership opportunity to define how agentic AI is built, governed, and scaled across a global bank.
Responsibilities
Agentic AI Platform & Agent Orchestration (Build from Scratch)
- Architect and build the enterprise agent orchestration platform from the ground up: agent runtime, planner/executor loops, tool and MCP-style integration layer, memory and context management, state persistence, and human-in-the-loop checkpoints
- Design multi-agent patterns — supervisor/worker hierarchies, task decomposition, delegation, hand-off, and recovery — with deterministic fallbacks where autonomy is not appropriate
- Build the tool integration surface that lets agents act safely on real enterprise systems: source control, CI/CD, ticketing, service catalogues, data platforms, and internal APIs, each with scoped, least-privilege credentials
- Define the evaluation and observability layer: golden datasets, offline and online evals, regression gates on agent behaviour, prompt and model version control, token and cost governance, and end-to-end tracing of every agent decision and action
- Establish reusable agent patterns, SDKs, and templates so other engineering teams can build compliant agents without rebuilding the plumbing
Enterprise Rollout & Adoption
- Own the enterprise rollout strategy: define the autonomy maturity ladder (assisted → supervised → autonomous), promotion criteria between stages, and the team onboarding path
- Prioritize and land high-value agentic use cases across the SDLC — code and change automation, release readiness, change-risk assessment, control evidence generation, incident triage, and developer self-service
- Define adoption metrics and prove impact with data: toil eliminated, cycle time, change failure rate, control coverage, cost per outcome
- Drive change management and enablement — playbooks, guardrail documentation, office hours, champion networks — so adoption scales beyond early adopters
AI Autonomy in a Highly Regulated Environment
- Design the control model for agent autonomy: authorization boundaries, approval and dual-control gates, blast-radius limits, kill switches, deterministic rollback, and segregation of duties between agents and humans
- Ensure every agent action is attributable, reproducible, and auditable — full decision and action lineage, immutable audit trails, and evidence artefacts that satisfy internal audit, risk, and regulators
- Partner with Risk, Compliance, Model Risk Management (MRM), Information Security, and Legal to move agentic use cases through model governance, risk acceptance, and control attestation
- Enforce data residency, entitlement inheritance, PII and confidential-data handling, prompt-injection and data-exfiltration defences, and secure secrets management across every agent path
- Codify controls as policy-as-code / compliance-as-code so they apply automatically rather than through manual review
Engineering Leadership
- Lead, mentor, and grow a team of AI and platform engineers — fostering technical excellence, psychological safety, and continuous improvement
- Collaborate with cross-functional stakeholders — product managers, security engineers, platform teams, controls partners, business analysts — to align the agent platform with organizational goals
- Drive cultural change by championing AI-augmented engineering practices and critical, creative thinking about controls and autonomy
Platform & Systems Engineering
- Design and build scalable, production-grade backend systems in Python and/or Golang, integrating with cloud-native and containerized infrastructure (Kubernetes / OpenShift)
- Ensure all platforms are secure, observable, and compliant with the firm's DevSecOps and SDLC requirements
- Champion best practices in system design including distributed systems, event-driven architectures, micro-services, and API-first design
- Keep the agent platform anchored in strong delivery foundations — CI/CD, progressive delivery (blue/green, canary), and everything-as-code
What We're Looking For
Leadership & Mindset
- A strategic thinker with a hands-on approach — comfortable moving between platform architecture and deep technical delivery
- A growth mindset with genuine passion for agentic AI, automation, and engineering innovation
- Strong communication and influencing skills — able to carry both engineering teams and senior risk/control stakeholders
- An advocate for inclusion, diversity, and psychological safety in all its forms
- A self-starter who thrives in ambiguity, brings structure to complexity, and energises the people around them
Ways of Working
- Pragmatic and risk-aware, with a creative approach to hard engineering problems — able to ship autonomy incrementally
- Committed to continuous improvement and data-driven, evidence-backed decision-making
- Enthusiastic about knowledge sharing, mentoring, and building high-performing teams
Experience & Skills
Required
- Proven experience leading AI/ML or platform engineering teams in product-focused environments
- Strong hands-on engineering experience in Python and/or Golang — building and shipping production systems
- Demonstrated experience building agentic AI systems in production: multi-agent orchestration, planning and tool-use loops, function calling, RAG pipelines, and frameworks such as LangChain / LangGraph, LlamaIndex, or equivalent
- Experience taking an AI platform from zero to enterprise scale — greenfield design, multi-tenant platform thinking, and driving adoption across many engineering teams
- Deep knowledge of LLM engineering: prompt and context engineering, evaluation harnesses, guardrails, model selection and routing, latency and cost management
- Experience operating AI systems under strict governance — auditability, human-in-the-loop controls, entitlements, data protection
- Proven experience with distributed systems, event-driven architectures, container-based micro-services, and cloud-native infrastructure (Kubernetes / OpenShift)
- Solid grounding in CI/CD and SDLC controls, with tools such as Harness, GitHub Actions, Tekton, or ArgoCD
- Hands-on experience with DevSecOps practices and security-by-design principles
Preferred
- Experience in regulated industries (financial services, fintech) with direct exposure to audit, model risk management, controls, or governance
- Exposure to policy-as-code and compliance-as-code frameworks (e.g., OPA, Rego)
- Familiarity with LLM observability and evaluation tooling, and observability stacks such as Grafana, OpenTelemetry, Loki, Prometheus
- Experience with MLOps platforms and model deployment pipelines (e.g., MLflow, Kubeflow, SageMaker, Vertex AI)
- Familiarity with agent interoperability standards (e.g., MCP, agent-to-agent protocols)
- Background in agile, product-oriented engineering teams with a continuous delivery mindset
------------------------------------------------------
Job Family Group:
Technology------------------------------------------------------
Job Family:
Systems & Engineering------------------------------------------------------
Time Type:
Full time------------------------------------------------------
Primary Location:
Tampa Florida United States------------------------------------------------------
Primary Location Full Time Salary Range:
$141,440.00 - $212,160.00
In addition to salary, Citi’s offerings may also include, for eligible employees, discretionary and formulaic incentive and retention awards. Citi offers competitive employee benefits, including: medical, dental & vision coverage; 401(k); life, accident, and disability insurance; and wellness programs. Citi also offers paid time off packages, including planned time off (vacation), unplanned time off (sick leave), and paid holidays. For additional information regarding Citi employee benefits, please visit citibenefits.com. Available offerings may vary by jurisdiction, job level, and date of hire.
------------------------------------------------------
Most Relevant Skills
Please see the requirements listed above.------------------------------------------------------
Other Relevant Skills
For complementary skills, please see above and/or contact the recruiter.------------------------------------------------------
Anticipated Posting Close Date:
Sep 22, 2026------------------------------------------------------
Automated Processing and AI
We use automated processing, including artificial intelligence, for our legitimate business interests (or our reasonable and appropriate business purposes) to identify and align the candidate's skills and abilities with a specific job opening. Additionally, if you so choose, or consent, we can match your skills and abilities to other suitable roles at Citi.
Importantly, all our hiring processes and decisions, including determining your suitability for a role, are conducted, checked, and decided by individuals. Our automated processing and AI do not involve relying on automatic or autonomous decision-making. Please refer to any Jurisdictional Considerations, with specific provisions for your country (where relevant) for further details.
Illinois residents – AI Notice and Right
------------------------------------------------------
Citi is an equal opportunity employer, and qualified candidates will receive consideration without regard to their race, color, religion, sex, sexual orientation, gender identity, national origin, disability, status as a protected veteran, or any other characteristic protected by law.
If you are a person with a disability and need a reasonable accommodation to use our search tools and/or apply for a career opportunity review Accessibility at Citi.
View Citi’s EEO Policy Statement and the Know Your Rights poster.