S

State Street

Holds assets for institutions and runs index funds

Generative AI Platform Manager

Engineering LeadershipConfirmed live in the last 24 hours
Full-timeIn-person

About the Job

Requirements

  • Primary depth in AWS Bedrock, Azure AI Foundry, Azure OpenAI, and Databricks is required, with strong skills in infrastructure-as-code authorship, Harness continuous integration and continuous delivery, FinOps, and AI-assisted developer tooling.
  • Bedrock knowledge including Knowledge Bases, Agents, Guardrails, Flows, provisioned throughput, Claude, Nova, Llama, Mistral, retrieval-augmented generation and vector stores such as OpenSearch, pgvector, Pinecone, and Kendra, plus agent frameworks such as Bedrock Agents, LangGraph, and Strands.
  • Azure AI knowledge including Foundry hubs and projects, prompt flow, Agent Service, Azure OpenAI, GPT-4o and GPT-4.1, o-series models, provisioned throughput units, AI Search, and Content Safety.
  • Databricks knowledge including Mosaic AI, Vector Search, AI Gateway, MLflow, Unity Catalog, Asset Bundles, and Delta Lake.
  • Infrastructure-as-code skills with CDK L2/L3 in TypeScript or Python, Bicep, multi-cloud Terraform, and secrets management.
  • Continuous integration and delivery skills with Harness YAML pipelines, delegates, templates, GitOps, OPA/Rego, GitHub Actions, and Azure DevOps.
  • FinOps skills in token, provisioned-throughput-unit, and Databricks-unit governance, tagging, chargeback, capacity planning, and token/model optimization.
  • Developer tooling skills with Claude Desktop, MCP server authoring, Claude Code, Copilot, and Cursor.
  • Deep AWS and Azure knowledge, including networking, IAM/Entra ID, private endpoints, Docker/Kubernetes (EKS/AKS), serverless, and API design; complementary GCP/Vertex AI knowledge.
  • Data skills involving Delta Lake, Unity Catalog, retrieval-augmented-generation and fine-tuning data pipelines, governance, and lineage.
  • Observability skills with CloudWatch, Azure Monitor, App Insights, Grafana, Arize, and Datadog.
  • Python as the primary language and TypeScript.
  • Experience scaling senior teams hands-on, executive communication, and owning budgets and roadmaps.
  • A bachelor's or master's degree in Computer Science, Engineering, or IT.
  • At least 10 years of platform, cloud, or site reliability engineering experience.
  • At least 5 years of hands-on AWS and Azure experience, with production depth in Bedrock, Azure OpenAI/Foundry, and Databricks.
  • At least 3 years of technical leadership or people management experience.

Responsibilities

  • Lead the design and delivery of the enterprise Generative AI platform.
  • Own enterprise Bedrock strategy, including model access governance, provisioned throughput, cross-region inference, and multi-account architecture.
  • Operationalize Bedrock Knowledge Bases, Agents, Guardrails, Prompt Management, Flows, and Model Evaluation.
  • Lead the foundation-model lifecycle across Claude, Nova/Titan, Llama, Mistral, and Cohere.
  • Design retrieval-augmented generation on Bedrock with OpenSearch Serverless, Aurora pgvector, and Kendra.
  • Optimize consumption through on-demand versus provisioned throughput, model routing, prompt caching, and token efficiency.
  • Own Foundry hub/project architecture, model catalog governance, and quota management.
  • Lead Azure OpenAI patterns, including provisioned throughput units versus pay-as-you-go, capacity planning, and content filters, across GPT-4o, GPT-4.1, and o-series models.
  • Architect retrieval-augmented-generation and agent workloads with AI Search, prompt flow, and Agent Service.
  • Implement Content Safety, private networking, Entra ID, customer-managed keys, and data residency controls.
  • Own Databricks strategy on AWS and Azure, including workspace architecture, Unity Catalog, and cluster policies.
  • Lead Mosaic AI adoption across Model Serving, Vector Search, Feature Store, AI Gateway, and MLflow.
  • Architect fine-tuning/pretraining pipelines and lakehouse-native retrieval-augmented generation on Delta Lake.
  • Establish Databricks-unit cost controls and serverless governance.
  • Author L2/L3 CDK constructs in TypeScript/Python, Bicep modules, and Terraform for multi-cloud AI infrastructure.
  • Codify secure-by-default golden paths that enable teams to launch Generative AI workloads in hours.
  • Standardize configuration-as-code, secrets management, and drift detection.
  • Own Harness pipelines, templates, and delegates for Bedrock, Azure OpenAI, Foundry, and Databricks Asset Bundles.
  • Integrate GitOps, Terraform/CDK/Bicep, OPA policy-as-code, and approval gates.
  • Establish reference pipelines with linting, security scanning, model evaluation, and cost checks.
  • Champion Claude Desktop, MCP servers, and AI coding assistants including Copilot, Cursor, and Claude Code.
  • Build internal MCP servers that expose enterprise systems to agentic clients.
  • Define secure usage patterns for regulated environments and measure productivity impact.
  • Establish security, compliance, and responsible-AI controls, including red-teaming, audit logging, and guardrails.
  • Build observability across CloudWatch, Azure Monitor, Databricks system tables, Grafana, and Datadog.
  • Partner with Data, MLOps, Security, and Application leadership; recruit and mentor a team.

Desired Qualifications

  • Financial-services or regulated-industry experience is strongly preferred.
  • Experience publishing CDK construct libraries, Bicep registries, or Terraform modules consumed enterprise-wide.
  • Enterprise-scale Harness pipeline authoring and rollout experience.
  • Fine-tuning or continued-pretraining experience using LoRA, QLoRA, RLHF, or DPO on Mosaic AI, Azure OpenAI, or Bedrock Custom Models.
  • Model optimization experience with quantization, distillation, prompt caching, or speculative decoding.
  • Provisioned-throughput-unit/provisioned-throughput capacity planning at scale.
  • Production MCP servers and Claude Desktop deployments in regulated environments.
  • Experience with agentic frameworks such as Bedrock Agents, Azure AI Agent Service, Databricks Agent Framework, LangGraph, or Strands.
  • Experience with AI safety/guardrail frameworks such as Bedrock Guardrails, Azure Content Safety, NeMo, or Guardrails AI.
  • Measurable FinOps outcomes, including token, Databricks-unit, or inference-cost reduction.
  • Open-source contributions to CDK, Terraform providers, or MCP.
  • Experience with multi-tenant AI platforms and chargeback/showback.
  • Board- or regulator-level presentation experience.
  • One or more preferred certifications: AWS Solutions Architect Professional, ML Specialty, or DevOps Engineer Professional; Azure AZ-305, AI-102, AZ-400, or OP-100; Databricks Data Engineer Professional, ML Professional, or Generative AI Engineer Associate; GCP Professional Cloud Architect or ML Engineer; Harness Certified Expert (CD or Platform); or Anthropic Claude Builder or partner credentials.

About the Company

State Street

Holds assets for institutions and runs index funds

State Street holds and services investments for pension funds, insurers, fund companies, and other large institutions, and it also runs index funds through State Street Investment Management. It launched the first US-listed exchange-traded fund, the SPDR S&P 500 ETF Trust, in 1993, and in 1924 it became custodian of the first US mutual fund. As a custody bank, it safekeeps clients' stocks and bonds, completes their trades, calculates what their funds are worth, and lends their securities to other investors. It earns fees on the assets it services and manages, plus income from currency trading and from interest on client cash. State Street traces its roots to a bank chartered in 1792 and is based in Boston, Massachusetts.

Company size
10,001+
Company stage
IPO
Headquarters
Boston, MA
Founded
1792

Recently Posted Jobs

Explore more opportunities at State Street

See more jobs