Senior Machine Learning Engineer

Job Summary

Machine Learning Engineers work to deploy end-to-end solutions to business problems leveraging AI and/or ML principles as needed to create those solutions. MLEs will take requests from stakeholders, define the components required for the project, gather data necessary for project EDA and training, then work with stakeholders to develop a plan around the productionized use of the solution, and work to put that solution into final production.

Responsibilities

• Consult with stakeholders to gather business requirements, translate them into agentic AI and data solutions, design high-level agent and model architectures, and demonstrate deep expertise in advanced analytics, LLMs, and AI/ML techniques to design, prototype, and build production-grade solutions to business problems.
• Architect, build, and deploy agentic AI systems (single-agent and multi-agent workflows) on Google Cloud, leveraging Google's Customer Engagement Suite (CES), Vertex AI Agent Builder, and Gemini-family models to automate enterprise workflows in customer engagement, sales, marketing, and operations.
• Design, integrate, and orchestrate the tools, APIs, function calls, retrieval pipelines (RAG), memory stores, and guardrails that extend agent capabilities, and own the end-to-end deployment, observability, evaluation, and lifecycle management of these agents in production.
• Lead communication with other stakeholders to drive agentic use case development and manage expectations on model and agent limitations, latency, cost, and lead times.
• Analyze data to identify useful relations, patterns and features that are predictive of user behaviors, preferences, intents, and interests, and use these signals to ground and personalize agent behavior.
• Manage and execute entire projects from start to finish, including cross-functional project management; data collection and manipulation, analysis and modeling; communication of insights and recommendations; productionalization of final model and agent products.
• Share findings with stakeholders to improve business decisions and/or influence strategic direction.
• Monitor and stay updated with industry trends and emerging technologies in agentic AI, foundation models, and MLOps/AgentOps to identify opportunities for innovation and improvement.
• Develop and maintain end-to-end modeling and agent code, and standardize the code for reusability in the production environment.
• Profile users including customer segmentation to help the marketing team target specific audiences for upgrading services and for user retention, and operationalize these insights through agent-driven engagement.

Qualifications

• Degree in a quantitative discipline, such as Data Science, Applied Mathematics, Statistics, Economics, Operations Research, Computer Science, Mathematics, Physics, Biology, Chemistry or Engineering. An advanced degree, Data Science bootcamp or MOOC certification is a plus
• 3-5 years of work experience in classification, regression, clustering, natural language processing (NLP), experiments, and optimization
• Hands-on experience with Google's Customer Engagement Suite (CES) is required and non-negotiable, including building, configuring, and deploying solutions across CES components (e.g., Conversational Agents / Dialogflow CX, Agent Assist, Conversational Insights) for enterprise customer engagement use cases
• Demonstrated experience building agentic AI systems in production - including single-agent and multi-agent architectures, planning and reasoning loops, tool/function calling, and orchestration with frameworks such as Vertex AI Agent Builder, ADK (Agent Development Kit), LangGraph, LangChain, CrewAI, or AutoGen
• Proven ability to integrate new tools and external systems (REST/GraphQL APIs, internal microservices, databases, vector stores, knowledge bases, MCP servers) to extend agent capabilities, and to own deployment, CI/CD, monitoring, evaluation, and guardrails for agents running in production
• Ability to apply Bayesian inference, frequentist statistics, causal modeling, and/or machine learning techniques
• Experience with any of these: customer segmentation, A/B experiments, quasi-experiments, sales forecasting, churn propensity modeling, customer lifetime value analysis, credit risk, geospatial analytics, survey key-drivers, marketing mix modeling, multi-touch attribution, or recommender systems
• Highly skilled in R and Python for statistical and machine learning programming
• Highly skilled in SQL & Python coding to wrangle and explore structured & unstructured data
• Proficient with server or Cloud computing platforms, such as Google Compute Engine or EC2
• Proficient with data warehouses, such as Oracle, BigQuery, or AWS
• Subject matter scientist that can review the literature to identify state-of-the-art solutions to a business problem

Preferred Qualifications

• Google Cloud certifications (e.g., Professional Machine Learning Engineer, Professional Cloud Architect, or Generative AI Leader) and demonstrated specialization in Google's CES and Vertex AI ecosystems
• Experience with Gemini models, function calling, structured outputs, prompt engineering, prompt evaluation, and fine-tuning / parameter-efficient tuning of foundation models on Vertex AI
• Experience designing Retrieval-Augmented Generation (RAG) pipelines with vector databases (e.g., Vertex AI Vector Search, Pinecone, Weaviate, pgvector) and grounding agents on enterprise knowledge
• Experience with AgentOps and LLMOps tooling - tracing, evaluation harnesses, online/offline evals, red-teaming, prompt and tool versioning, cost and latency observability
• Experience implementing responsible AI practices for agents - safety, PII handling, hallucination mitigation, human-in-the-loop review, and policy/guardrail enforcement
• Experience integrating agents with enterprise systems such as CRM (Salesforce), CCaaS / contact center platforms, billing, ticketing, and identity providers in a regulated environment
• Experience with containerization and orchestration (Docker, Kubernetes / GKE, Cloud Run) and infrastructure-as-code (Terraform) for deploying agentic services
• Contributions to open-source agentic AI projects, publications, patents, or conference talks in the GenAI / agentic AI space