Senior Engineer - Agentic AI, Assistant Manager

Who we are looking for We are looking for a hands-on Senior AI Engineer with strong software engineering experience to build and deploy secure, production-grade AI agents.

The ideal candidate has experience with agent frameworks, orchestration, enterprise integrations, evaluation and guardrails, AWS Bedrock or equivalent cloud services, and modern deployment practices.

Knowledge of responsible AI and governance is essential; experience with Knowledge Graphs, RAG, or GraphRAG is an advantage.

Why this role is important to us This role is central to building secure, scalable, and production-grade AI capabilities that improve how enterprise processes are delivered.

By engineering intelligent agents, multi-agent workflows, enterprise integrations, evaluation frameworks, guardrails, and cloud-native deployment patterns, the Senior AI Engineer will help automate complex work, improve operational efficiency, and accelerate reliable AI adoption.

The role also strengthens enterprise knowledge and reasoning through complementary knowledge graph capabilities while ensuring solutions meet responsible AI, security, governance, auditability, and production-support requirements.

About the team — Alpha Intelligence Alpha Intelligence is State Street’s AI & Automation Center of Excellence for Alpha Implementations — the team rewiring how the industry’s first front-to-back platform gets delivered.

We turn the deep expertise locked inside our implementation teams into engineered processes, governed knowledge and intelligent agents that cut manual effort, compress client go-live timelines and make every implementation faster and more predictable than the last.

We operate as one global team across the North America, EMEA, and APAC, combining process engineering, automation delivery, AI platform engineering and responsible-AI governance under a single roof.

We work on real client implementations, not pilots that sit on a shelf — what we build is adopted, measured and scaled.

If you want to shape how AI is applied to one of the most complex delivery landscapes in financial services, this is the team to do it in.

What you will be responsible for This role will primarily focus on designing, engineering, deploying, and operating secure, production-grade Agentic AI capabilities, with a complementary focus on knowledge graph and Graph RAG solutions that improve enterprise grounding, retrieval, and reasoning.

Design, develop, and integrate AI agents with enterprise APIs, applications, tools, data sources, knowledge repositories, and workflow services.

Build reusable agent workflows using Lang Chain, Lang Graph, Llama Index, Semantic Kernel, or equivalent frameworks, applying modular and maintainable engineering patterns.

Engineer single-agent and multi-agent orchestration, including planning, routing, memory, state management, tool or function calling, retries, checkpointing, and failure recovery.

Establish agent evaluation and observability using Lang Smith or equivalent tools, including test datasets, quality metrics, execution traces, regression testing, and production monitoring.

Implement human-in-the-loop approvals, agent guardrails, structured evaluation, and responsible execution controls for sensitive, high-impact, or irreversible actions.

Configure and integrate AWS Bedrock Agents, Guardrails, foundation models, and equivalent cloud-native agent services.

Package and deploy agent services using containers, serverless patterns, infrastructure as code, CI/CD pipelines, and enterprise cloud security controls.

Apply Responsible AI, data privacy, access control, auditability, model-risk, and AI-governance requirements across design, testing, deployment, monitoring, and production support.

Contribute to ontology and knowledge graph design, graph data modeling, ingestion, entity and relationship enrichment, lineage, provenance, and semantic validation using enterprise graph technologies.

Build supporting Python APIs and RAG or Graph RAG components, including document processing, hybrid retrieval, graph traversal, grounding, source attribution, citations, testing, troubleshooting, performance tuning, documentation, and cross-functional delivery.

What we value The successful candidate must have excellent verbal, written, and presentation skills and be able to communicate complex AI, graph, data, and architecture concepts to engineering teams, business stakeholders, and senior technology leaders.

The candidate should demonstrate strong ownership, structured problem solving, pragmatic architecture, reusable engineering, collaboration, and a consistent focus on secure, reliable, measurable, and governed enterprise AI outcomes.

Must-have skills Experience developing or integrating AI agents with enterprise APIs, tools, data sources, knowledge repositories, and workflow services.

Experience with agent frameworks such as Lang Chain, Lang Graph, Llama Index, Semantic Kernel, or equivalent.

Experience with Lang smith and AI Eval tools, or equivalent.

Exposure to single-agent and multi-agent orchestration, tool or function calling, agent memory, planning, routing, and state management.

Experience implementing human-in-the-loop controls, agent guardrails, structured agent evaluation, and responsible execution patterns.

Experience with AWS Bedrock Agents, Guardrails, foundation model integrations, or equivalent cloud-native agent services.

Experience with containerization, serverless deployment, infrastructure code, or enterprise cloud security patterns.

Knowledge of responsible AI, data privacy, access controls, auditability, model risk, and AI governance in regulated environments.

Nice-to-have skills Hands-on experience designing ontologies, semantic models, metadata structures, taxonomies, entity types, and relationship definitions.

Strong experience with at least one enterprise graph platform such as Neo4j, Amazon Neptune, Graph DB, or equivalent technology.

Proficiency in graph data modeling and query languages such as C