Sr. Principal, Product Management

Smarsh is the world’s leading provider of intelligence on sensitive communications for regulated industries. They are seeking a product leader to own MLOps and Agentic strategy for Smarsh’s Analytics Platform, focusing on model governance, risk management, and cross-functional collaboration.

Responsibilities:

• Define and socialize the multi-year MLOps platform roadmap, spanning model training, evaluation, deployment, monitoring, and retirement across global regions
• Architect product solutions for multi-region model inference with data residency constraints — ensuring our models operate compliantly across jurisdictions without sacrificing performance
• Drive model observability and drift detection capabilities from concept to adoption, in partnership with Data Science and Engineering
• Own the vendor and tooling strategy across the MLOps stack evaluating build vs. buy tradeoffs with a clear lens on total cost of ownership and compliance posture
• Publish the Agent Framework Reference Architecture defining how autonomous and semi-autonomous agents are designed, tested, and deployed within a regulated FinServ environment
• Lead product strategy for agentic workflow reliability, including human-in-the-loop design patterns, tool-use governance, and failure mode handling
• Define and document the Bring Your Own Model establishing how the organization governs the introduction of external models and custom prompts while maintaining auditability and control
• Translate emerging agent framework patterns (e.g., multi-agent orchestration, RAG pipelines, memory and context management) into concrete product requirements and phased delivery plans
• Champion model governance as a product capability not a compliance checkbox by embedding risk controls directly into the platform so teams can move fast within guardrails
• Drive adoption of the Model Governance Framework with Model Risk Management (MRM) teams, ensuring our AI systems meet regulatory expectations across all deployment regions
• Define meaningful KPIs and observability standards that allow risk and compliance teams to assess model health without becoming a bottleneck to innovation
• Develop and maintain a model risk taxonomy that scales across foundation models, fine-tuned models, and agentic systems
• Partner with Applied ML, Engineering, Product, Finance, and Procurement to ensure alignment on strategy, performance, and cost
• Represent the MLOps and Agent Frameworks product domain to senior leadership, translating complex technical tradeoffs into strategic narratives that drive decision-making
• Establish and govern the product council or working group that coordinates AI platform decisions across business units and geographies
• Collaborate with product teams to embed analytics consistently into their workflows

Requirements:

• 8+ years of product management experience, with at least 3 years focused on ML infrastructure, MLOps platforms, or AI/ML product strategy
• Experience partnering directly with Data Science/Applied ML and Engineering teams
• Demonstrated experience shipping ML platform or AI infrastructure products in a regulated industry — financial services, healthcare, or similarly complex environments strongly preferred
• Hands-on experience/ strong familiarity with MLOps tooling (e.g., MLflow, Kubeflow, SageMaker, Vertex AI, Databricks) and the ability to evaluate them critically as a product strategist
• Working knowledge of LLM orchestration frameworks (e.g., LangChain, LlamaIndex, AutoGen, CrewAI) and the architectural tradeoffs they represent
• Demonstrated experience managing build vs. buy decisions or large-scale vendor evaluations
• Excellent communication skills and ability to work across product, engineering, and commercial teams
• Willingness to support deal cycles with competitive insights and analytics expertise
• Experience with regulated industries or enterprise SaaS
• Engineering or data science background
• Ability to engage credibly in architecture reviews — you do not need to write the code, but you need to understand the implications of the decisions being made
• Fluency in data residency concepts, multi-region cloud deployment patterns, and the engineering complexity they introduce for global AI products
• Understanding of model risk management (MRM) frameworks and how they apply to modern ML and LLM systems in financial services
• Comfort with ambiguity in emerging areas — Bring your own model strategy, agentic reliability, and multi-agent orchestration are fields where the playbook is still being written