New opportunity

AI Enablement Lead

Waymo · Mountain View, United States

About this role

Waymo’s AI Enablement Lead helps the Data Intelligence team deploy AI across general and administrative functions. The role combines technical leadership with hands-on prototyping and enterprise integration to automate business processes and support data-driven decisions. It follows a hybrid schedule in Mountain View and reports to the Data Intelligence Manager.

Responsibilities

Serve as an AI domain authority across G&A: mentor team members, establish technical standards, and advise leadership on investments and roadmap priorities.
Build proofs of concept using large language models, agent frameworks, and workflow orchestration tools. Review code and prompts, guide system design, and define evaluation methods and performance benchmarks.
Work with Finance, HR, Sales, Supply Chain, and Marketing to identify manual bottlenecks and automate complex processes. Oversee connections between AI agents and enterprise systems through APIs, data pipelines, and workflow platforms.

Required qualifications

BS in Computer Science or a related technical field, or equivalent practical experience.
8–10 years of professional experience across data science, data engineering, and business intelligence, plus 3+ years of technical experience building and shipping AI/ML products.
Hands-on experience using GenAI tools to turn prompts into small prototypes in a constrained environment and use those prototypes to validate requirements.
Strong Python proficiency; experience with AI/ML frameworks such as LangChain, Vertex AI Agent Builder, TensorFlow, and PyTorch; and experience with cloud platforms, GCP, Vertex AI, A2A, MCP, and ADK in enterprise systems. Ability to explain technical AI concepts to nontechnical stakeholders.

Preferred qualifications

Secure integration of AI and LLM solutions with enterprise SaaS platforms such as Salesforce, SAP, Workday, and ServiceNow; advanced agentic patterns including multi-agent orchestration, complex RAG architectures, and dynamic tool use.
Experience influencing cross-functional leaders, managing organizational change and adoption, and applying AI governance, data privacy, RBAC, and data-loss-prevention practices to internal AI applications. Ability to lead projects through ambiguity from conception to launch.

The expected US base salary range is $200,000–$247,000. Eligible employees may participate in a discretionary annual bonus program, equity incentive plan, and company benefits program.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

Generative AILarge Language ModelsAI agentsPythonLangChainVertex AI Agent BuilderTensorFlowPyTorchGoogle Cloud PlatformVertex AIA2AMCPADKWorkflow orchestrationAPI integrationData pipelinesPrompt engineeringAI evaluationData scienceData engineeringBusiness intelligenceRAGMulti-agent orchestrationSalesforceSAPWorkdayServiceNowAI governanceData privacyRole-based access control (RBAC)

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Set AI direction across G&A
Serve as the AI domain authority: mentor team members, establish technical standards, and advise leadership on investments and roadmap priorities.
Prototype to validate requirements
Build LLM- and agent-based proofs of concept with workflow orchestration. The posting specifically requires using GenAI tools to create small prototypes in a constrained environment and use them to validate requirements.
Define quality and performance
Review code and prompts, guide system design, and establish evaluation methods and performance benchmarks for AI solutions.
Find valuable business workflows
Partner with Finance, HR, Sales, Supply Chain, and Marketing to identify manual bottlenecks and apply AI to complex processes; explain technical choices to nontechnical stakeholders.
Connect agents to enterprise systems
Oversee deployment through APIs, data pipelines, and workflow platforms. The posting also calls for enterprise application of GCP, Vertex AI, A2A, MCP, and ADK; secure SaaS integration is preferred.

Likely challenges to think through

Speed versus production evidence
Preparation hypothesis: quick prototypes may reveal requirements without proving that a workflow is ready to deploy. Prepare an example of how you would move from a constrained prototype to evaluation benchmarks, design review, and a deployment decision.
Useful automation versus controlled access
Preparation hypothesis: agents spanning business systems may need data access while handling sensitive internal information. Prepare to explain trade-offs among workflow coverage, permissions, privacy, RBAC, and data-loss prevention; these controls appear in the preferred qualifications.
Shared roadmap versus varied business needs
Preparation hypothesis: bottlenecks across multiple G&A functions may compete for investment, while adoption depends on nontechnical teams. Prepare a way to prioritize use cases and explain trade-offs to leaders without assuming every function needs the same solution.

Practice your answers

Technical practice

How would you identify a G&A process worth automating with AI?

What to coverSuggested preparation: Explain how you would map a manual bottleneck with a business partner, then assess data availability, process complexity and expected value. Distinguish a task suited to an LLM or agent from one better handled by conventional workflow automation.

How would you use a small GenAI prototype to validate requirements?

What to coverSuggested preparation: Describe a constrained prototype built from prompts and how you would use the artifact to test assumptions with users. Address what the prototype could establish, what it could not, and how you would decide whether to proceed.

How would you design an LLM-powered, multi-step business workflow?

What to coverSuggested preparation: Sketch the responsibilities of the LLM, agent framework and workflow orchestrator, including where tool calls and approvals would occur. Explain how you would make failures visible and prevent an incomplete step from silently affecting later steps.

When would you use multi-agent orchestration or dynamic tool use?

What to coverSuggested preparation: Compare a single agent with coordinated agents for a cross-platform workflow, weighing complexity against clearer separation of tasks. Discuss how you would constrain tool selection and assess whether approaches involving A2A, MCP or ADK add value in an enterprise system.

How would you evaluate whether a business workflow needs RAG?

What to coverSuggested preparation: Explain when retrieval from business information could improve an LLM response and when it would add unnecessary complexity. Cover source quality, access permissions and how you would test whether retrieved content improves results.

What would you examine in a review of Python code and prompts for an AI proof of concept?

What to coverSuggested preparation: Discuss correctness, maintainability, prompt behavior and handling of unexpected inputs. Explain how review standards would differ between a quick requirements-validation prototype and a system being prepared for deployment.

How would you define evaluation methods and performance benchmarks for an internal AI application?

What to coverSuggested preparation: Identify measures tied to the business task, such as output quality, task completion, latency and human review burden. Explain how you would assemble representative cases, inspect failures and set a decision threshold before expanding use.

How would you troubleshoot an agent’s connection to an enterprise system?

What to coverSuggested preparation: Walk through isolating an issue across the agent, API, data pipeline and workflow platform while preserving traceability. For a SaaS integration such as Salesforce, SAP, Workday or ServiceNow, address permissions, data consistency and safe recovery from failed actions.

How would you protect internal corporate data in an AI workflow?

What to coverSuggested preparation: Explain how role-based access control, data privacy practices and data-loss prevention would shape retrieval, prompts, tool access and outputs. Discuss how you would verify that an agent cannot expose or act on data beyond a user’s authorization.

How would you choose among AI frameworks and cloud services for a proof of concept?

What to coverSuggested preparation: Compare the fit of Python-based tooling, a framework such as LangChain, and GCP or Vertex AI services against the prototype’s requirements. Address integration needs, operational complexity and what evidence would justify moving from prototype to a supported deployment.

Behavioural practice

Tell me about a time you aligned a business team and technical team on an AI opportunity.

What to coverSuggested preparation: Use a real situation involving differing views of a process or its requirements; explain your actions to clarify the problem and communicate technical trade-offs to nontechnical stakeholders. State the result and how you knew the teams had reached a useful decision.

Describe a time you mentored engineers or raised technical standards.

What to coverSuggested preparation: Give a real example of the starting practices, the guidance or review standards you introduced, and how you supported others in adopting them. Explain the resulting change in team decisions or work quality without attributing outcomes you cannot substantiate.

Tell me about a time you influenced a technical investment or roadmap priority.

What to coverSuggested preparation: Set out the competing options and the evidence you brought to leadership, including benefits, risks and trade-offs. Describe your actions, the decision reached and what happened afterward.

Describe a time you helped a nontechnical group adopt a new tool or workflow.

What to coverSuggested preparation: Explain the users’ initial concerns or barriers, what you did to address them, and how you communicated the change. Give the result using observable adoption or workflow evidence where available.

Tell me about a project you led from an ambiguous idea toward launch.

What to coverSuggested preparation: Describe the uncertain starting point, how you narrowed scope and validated requirements, and how you coordinated decisions as the work progressed. Explain the outcome and one trade-off you made along the way.

Sharpen your knowledge

Topics to prepare

Enterprise agent architecture
Posting: The role calls for LLM, agent-framework and workflow-orchestration knowledge, including GCP, Vertex AI, A2A, MCP and ADK in enterprise systems. Suggested exercise: sketch an agent that handles a multistep G&A request; explain where orchestration, tool access and human review belong.
Evaluation and performance benchmarks
Posting: The lead defines evaluation methods and benchmarks and reviews prompts and system designs. Suggested exercise: define test cases, quality measures, failure categories and a benchmark for a proposed process-automation proof of concept.
Enterprise integration and data flow
Posting: The role oversees connections between AI agents and enterprise systems through APIs, data pipelines and workflow platforms. Suggested exercise: diagram the data flow for an agent that reads a business-system record and proposes an update, including where errors are handled.
G&A opportunity selection and technical leadership
Posting: The lead identifies bottlenecks with Finance, HR, Sales, Supply Chain and Marketing, sets standards, mentors others and advises on roadmap priorities. Suggested exercise: compare two hypothetical manual workflows by value, feasibility and risk, then explain the priority to a nontechnical leader.
Privacy, governance and advanced agent patterns
Posting: Governance, data privacy, RBAC and data-loss prevention are preferred, as are multi-agent orchestration, complex RAG and dynamic tool use. Suggested exercise: explain when a simple workflow is preferable to an advanced agent, and identify access and data-exposure risks in each design.

Skills to revise

Prompt-to-prototype validation
Posting: Hands-on use of GenAI tools to turn prompts into small prototypes in a constrained environment, then validate requirements with the artifact, is mandatory. Suggested exercise: build a small mock-data G&A workflow prototype and use it to record which requirements are confirmed or still unclear.
Python and prompt review
Posting: Strong Python proficiency and review of code and prompts are expected. Suggested exercise: write a short Python agent-workflow prototype, then review it for unclear prompts, failure handling and maintainability.
Agent and workflow integration
Posting: Proofs of concept use agent frameworks and orchestration tools; deployment involves APIs and data pipelines. Suggested exercise: connect a prototype to a mock enterprise API and trace inputs, outputs and failures. LangChain and Vertex AI Agent Builder are examples named in the posting, not required choices for this exercise.
Hands-on evaluation
Posting: The lead defines evaluation methods and performance benchmarks. Suggested exercise: run the same small test set through two prompt or workflow versions, record results and explain which version you would advance.
Stakeholder explanation and adoption
Posting: The role requires explaining AI concepts to nontechnical stakeholders; influencing leaders and supporting adoption are preferred. Suggested exercise: give a brief, jargon-free explanation of a prototype’s benefits, limitations and proposed rollout decision for a G&A audience.

Your preparation checklist

Prepare a shipped-product walkthrough
The role asks for experience building and shipping AI/ML products across data science, data engineering, and BI. Prepare one project narrative covering your contribution, architecture, launch decisions, measured outcomes, and what you would change.
Build a small, constrained GenAI prototype
The posting explicitly calls for turning prompts into small prototypes to validate requirements. Prepare a Python example with a stated constraint, a working artifact, and a short account of what it did—and did not—validate.
Sketch an enterprise-agent design
For practice, choose a hypothetical manual bottleneck in one listed function, such as Finance or Supply Chain. Draw the LLM or agent workflow, API and data-pipeline connections, human review points, and where relevant tools such as GCP, Vertex AI, A2A, MCP, or ADK would fit.
Define an evaluation plan
The role includes reviewing code and prompts and setting evaluation methods and benchmarks. Prepare a one-page plan with representative test cases, quality and performance measures, failure analysis, and criteria for revising a proof of concept.
Explain security and governance choices
Secure SaaS integration, privacy, RBAC, and data-loss prevention are preferred areas. Prepare an annotated design showing how you would limit agent access to internal data and connected systems; distinguish controls you have used from ones you propose.
Rehearse a leadership example
Prepare a concise account of how you set a technical standard, mentored others, or influenced a cross-functional roadmap through ambiguity. Include the decision trade-offs and a plain-language explanation for nontechnical stakeholders.
Bring questions about priorities and deployment
Ask the Data Intelligence Manager which G&A bottlenecks are highest priority, how prototypes are evaluated before deployment, who owns enterprise integrations and governance, and how adoption is measured.

Availability last checked: Nov 4, 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in United States. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…