Data Science Manager

Role summary We’re looking for a Manager, Data Scientist to lead the design and delivery of advanced AI/ML solutions—from problem framing and model development to deployment strategy, security, and cost management. You will manage a team of data scientists and collaborate closely with solution architects to ensure AI/ML efforts fit enterprise goals. You will direct model selection, validation, infrastructure integration, and governance to drive scalable, secure, and cost-effective AI services. What you’ll do • Lead end-to-end AI/ML project execution including predictive modeling, NLP, computer vision, and retrieval-augmented generation (RAG) initiatives. Define best practices, reference architectures, and operational patterns (real-time, batch, online/offline inference). • Oversee model lifecycle management: select appropriate LLMs/foundation models and classical machine learning algorithms, establish evaluation metrics (quality, latency, safety), and ensure guardrails and fallback mechanisms are in place. • Manage collaboration with cloud engineering teams on AWS AI/ML services (Amazon Bedrock, SageMaker Studio and Training, S3/Lake Formation, Kendra, OpenSearch, Lambda, EKS/ECS, etc.) for solution deployment and orchestration. • Guide vector search and retrieval system design, including embedding strategies, indexing, metadata filtering, and operational considerations using technologies like Kendra, Pinecone, OpenSearch, and others. • Drive development of RAG and agentic AI patterns, such as retrieval pipelines, prompt orchestration, tool integration, persona management, caching, and safety filtering. • Design and implement environment strategies across development, testing, and production, including GPU/accelerator resource allocation and CI/CD pipelines for models and prompts. • Balance cost optimization and performance tuning by forecasting resource needs, managing budgets, autoscaling, and employing techniques such as quantization and caching. • Enforce security, privacy, and compliance standards related to data governance, encryption, key management, and mitigation of prompt-injection or other threats. • Partner with ML engineers and data scientists on feature stores, experiment tracking, evaluation pipelines, and A/B testing to continuously improve model performance and reliability. • Establish operational readiness with defined SLOs/SLIs, telemetry, incident response plans, and capacity planning. • Mentor and develop your team, author documentation and best practices, and help enable cross-functional teams through training and reference solutions. Required qualifications • 8+ years experience in data science, AI/ML, and data engineering, with 3+ years managing teams and architecting production-scale AI solutions in enterprise environments. • Demonstrated leadership in delivering LLM-driven applications (chatbots, agents, RAG) and machine learning services (forecasting, classification, ranking) in production. • Deep familiarity with AWS AI/ML stack (Bedrock, SageMaker, S3, Glue, Redshift, Lambda, EKS, VPC, IAM, KMS). • Strong expertise in vector search technologies, embeddings, retrieval design, re-ranking, and metadata. Experience managing at least one managed vector store or search service. • Hands-on knowledge of MLOps best practices including CI/CD, model registries, automated evaluation, versioning, rollbacks, and observability. • Solid understanding of security, compliance, and governance for regulated data (PHI/PII), including encryption, access controls, and audit trails. • Strong financial acumen in managing compute and token budgets, optimizing latency, throughput, and cost in AI/ML workloads. • Excellent communication skills to lead teams and collaborate with stakeholders. Preferred qualifications • Experience with Azure AI and hybrid/multi-cloud architectures. • Familiarity with Snowflake Cortex/Databricks/MosaicML ecosystems. • Knowledge of agent frameworks (LangChain, LlamaIndex) and productionizing on cloud platforms. • Experience with event-driven streaming (Kafka/MSK), online feature stores, and real-time inference. • FinOps mindset with prior responsibility for large budgets involving GPU and token usage. • Background in regulated industries like healthcare or financial services; FedRAMP or equivalent compliance experience preferred. • Relevant cloud/AI certifications (e.g., AWS ML Specialty, AWS Solutions Architect Pro).