Job description

You care about how fast good models reach customers, and you know most of that time is lost after the science is done. Join us to make getting a model to production fast, repeatable, and safe. As a Vice President within the Personalization & Customer Insights (Intelligence) team, you help shape the vision for our modeling platform and own the roadmap for key parts of it. You are the voice of the modeler, the data scientists and ML engineers who build on the platform. You work with platform leadership and engineering to set direction, and you drive execution across Data Science, ML Engineering, Cloud and Infrastructure, Model Governance, and the product teams that depend on the models you serve. You are accountable for what your areas ship, on time and at quality. The platform serves models three ways: batch serving on a schedule, real-time serving in the moment a customer acts, and language model hosting for the small and large language models behind AI-written messages and assistants. Each comes with the reporting that makes it a product, not just infrastructure: performance, monitoring, explainability, cost and usage, service health, and the evidence model governance needs. The goal is speed to market. Today, getting a model into production is too often a custom project. This role helps make it the default path, so a trained model is served, monitored, and ready to test in days, not months. Job responsibilities • Help shape the platform strategy and own the roadmap for your areas of the model development lifecycle (MDLC), from deployment and serving to monitoring and governance. • Cut time to market: shorten the time from trained model to production and grow the share of models that go live without custom infrastructure work. • Own the roadmap for the serving paths in your scope (batch, real-time, and language model hosting), including the response time, uptime, and cost each one commits to. • Help make going live self-serve: one path to train, deploy, serve, and A/B test, with no ticket to another team, including for teams outside personalization. • Own the reporting that ships with your areas: performance, monitoring and drift, explainability, cost and usage, and service health, built for model owners, product stakeholders, and model governance. • For hosted language models, define what teams get out of the box: routing between self-hosted and external models, evaluation, tracing, guardrails, and cost reporting. • Prioritize against real constraints, balancing new capabilities with reliability, controls, regulatory needs, and technical debt. • Be accountable for delivery: your roadmap shipping to production, on time and at quality. • Partner with engineering on architecture and build-versus-buy decisions, staying close enough to the detail to pressure-test them. • Partner with Risk, Model Governance, and Compliance to build governance into the platform, automated where possible and trusted by reviewers. • Define and track KPIs for your areas (time to production, adoption, reliability, cost per inference, modeler satisfaction) and contribute to the business case for platform investment. Required qualifications, capabilities, and skills • 5+ years of product management experience or equivalent, including owning the roadmap for a technical or platform product. • Proven success delivering ML platform or MLOps products to production, in close partnership with ML Engineering, Data Science, and infrastructure teams. • Strong working knowledge of how models run in production: batch and real-time inference, deployment and versioning, monitoring and drift, and the latency, throughput, and cost tradeoffs of serving. • Experience with language models in production: hosting or serving small or large language models, routing, evaluation, and managing cost and response time per call. • A track record of measurably shortening the time it takes to get models or software into production (for example, through self-serve tooling, standard deployment paths, or automation). • Experience treating internal users as customers: user research, adoption metrics, documentation, and support that make a platform something teams choose to use. • Hands-on data fluency: able to explore and query data yourself (SQL/Python, or AI-assisted tools such as GitHub Copilot or Claude) to frame problems, check platform and model metrics, and move quickly. This role does not lead from a distance; familiarity with these tools is expected. • Accountable for delivery, not just direction: a track record of owning a product through to production and prioritizing under real constraints, balancing new work against operational stability, reliability, controls/regulatory needs, and technical debt. • Comfort with ambiguity: stepping into a platform that is still being built and helping decide what to standardize first. • Proven ability to drive execution across a highly matrixed organization; excellent stakeholder management and clear, structured written and verbal communication, including explaining technical tradeoffs to non-technical leaders. Preferred qualifications, capabilities, and skills • Experience building model monitoring, explainability, or observability products, including reporting for model risk management in a regulated industry. • Familiarity with common ML platform tooling (e.g., Databricks, MLflow, Ray, Kubernetes, AWS) and with GPU capacity planning. • Experience automating governance and documentation steps in the model lifecycle, including with AI agents. • Experience leading and developing product managers. • BS or MS in Computer Science, Engineering, Data Science, or a comparable field. • Familiarity with responsible AI, privacy, and fairness considerations for models and language models in financial services. Shape the vision and own the roadmap for how our platform serves ML models at scale — fast, reliable, and ready for production.

Requirements

Must-haves

You care about how fast good models reach customers, and you know most of that time is lost after the science is done. Join us to make getting a model to production fast, repeatable, and safe.
The goal is speed to market. Today, getting a model into production is too often a custom project. This role helps make it the default path, so a trained model is served, monitored, and ready to test in days, not months.
Help shape the platform strategy and own the roadmap for your areas of the model development lifecycle (MDLC), from deployment and serving to monitoring and governance.
Cut time to market: shorten the time from trained model to production and grow the share of models that go live without custom infrastructure work.
Own the roadmap for the serving paths in your scope (batch, real-time, and language model hosting), including the response time, uptime, and cost each one commits to.
Help make going live self-serve: one path to train, deploy, serve, and A/B test, with no ticket to another team, including for teams outside personalization.
Own the reporting that ships with your areas: performance, monitoring and drift, explainability, cost and usage, and service health, built for model owners, product stakeholders, and model governance.
For hosted language models, define what teams get out of the box: routing between self-hosted and external models, evaluation, tracing, guardrails, and cost reporting.
Prioritize against real constraints, balancing new capabilities with reliability, controls, regulatory needs, and technical debt.
Be accountable for delivery: your roadmap shipping to production, on time and at quality.
Partner with engineering on architecture and build-versus-buy decisions, staying close enough to the detail to pressure-test them.
Partner with Risk, Model Governance, and Compliance to build governance into the platform, automated where possible and trusted by reviewers.
Define and track KPIs for your areas (time to production, adoption, reliability, cost per inference, modeler satisfaction) and contribute to the business case for platform investment.
Required qualifications, capabilities, and skills
5+ years of product management experience or equivalent, including owning the roadmap for a technical or platform product.
Proven success delivering ML platform or MLOps products to production, in close partnership with ML Engineering, Data Science, and infrastructure teams.
Strong working knowledge of how models run in production: batch and real-time inference, deployment and versioning, monitoring and drift, and the latency, throughput, and cost tradeoffs of serving.
Experience with language models in production: hosting or serving small or large language models, routing, evaluation, and managing cost and response time per call.
A track record of measurably shortening the time it takes to get models or software into production (for example, through self-serve tooling, standard deployment paths, or automation).
Experience treating internal users as customers: user research, adoption metrics, documentation, and support that make a platform something teams choose to use.
Comfort with ambiguity: stepping into a platform that is still being built and helping decide what to standardize first.
Experience building model monitoring, explainability, or observability products, including reporting for model risk management in a regulated industry.
Familiarity with common ML platform tooling (e.g., Databricks, MLflow, Ray, Kubernetes, AWS) and with GPU capacity planning.

Nice-to-haves

Preferred qualifications, capabilities, and skills

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

mlaigosqlpythonkubernetesaws

Platform: oracle-cloud
Effort: medium

Benefits

Benefits detected in the listing.

Vision