Requisition ID # 173576
Job Category: Accounting / Finance
Job Level: Manager/Principal
Business Unit: Technology & Security
Work Type: Hybrid
Job Location: Oakland
Position SummaryWe are seeking a Principal Data Scientist
(AI Engineer focused) to serve as the chief technical staff member within Enterprise AI CoE. You will be the senior-most hands-on technical leader for enterprise AI execution, responsible for setting technical direction, designing scalable AI and agentic AI architectures, and guiding complex AI solutions from concept through production adoption. You will help lead significant changes in how PG&E adopts, governs, builds, and scales AI by establishing practical engineering standards, reusable solution patterns, and trusted technical approaches for generative AI, agentic systems, machine learning, and enterprise data science.
This is a highly visible technical leadership role requiring deep expertise in Python-based AI engineering, cloud AI platforms such as AWS Bedrock, agentic AI frameworks, retrieval-augmented generation, AI evaluation, observability, responsible AI, and production-grade MLOps/LLMOps. The successful candidate will work closely with Enterprise AI leadership, AI Cloud Engineering, Enterprise Architecture, Cybersecurity, Legal, Privacy, product teams, business sponsors, and line-of-business SMEs to ensure PG&E builds AI solutions that are technically sound, operationally scalable, secure, compliant, and tied to measurable business value.
This position is hybrid, working from your remote office and in-person at our Oakland Headquarters 1-2x per week or based on business needs or company requirements.
Job Responsibilities- Serve as the chief technical staff member for the Enterprise AI CoE, providing hands-on technical leadership across AI Advisory, AI Governance, and AI Factory priorities.
- Set the technical vision and architecture direction for enterprise AI execution, including generative AI, agentic AI, traditional machine learning, AI-enabled automation, and reusable AI solution patterns.
- Lead the design and implementation of production-grade AI systems using Python, AWS Bedrock, enterprise LLM gateways, agent orchestration frameworks, retrieval-augmented generation, vector search, APIs, and secure enterprise integrations.
- Establish reusable technical standards, reference architectures, design patterns, prompt and agent templates, evaluation harnesses, and engineering best practices that enable AI solutions to scale across PG&E.
- Act as the senior technical reviewer for complex AI use cases, model designs, agent workflows, AI architecture proposals, and production readiness decisions.
- Guide AI Factory execution by helping teams move from prototypes and pilots to reliable, monitored, governed, and supportable production AI solutions.
- Partner with business sponsors, product owners, data scientists, engineers, platform teams, and SMEs to translate high-value business needs into technically feasible and measurable AI solutions.
- Drive development of agentic AI capabilities, including task-based agents, single-domain agents, multi-agent workflows, tool use, memory, orchestration, human-in-the-loop controls, and enterprise system integration.
- Define and apply AI quality practices, including model and agent evaluation, regression testing, red teaming, responsible AI review, observability, performance monitoring, cost tracking, and lifecycle management.
- Mentor and coach data scientists, AI engineers, product teams, and business-facing practitioners to raise the technical maturity of AI delivery across PG&E.
- Represent the Enterprise AI CoE's technical perspective in cross-functional forums with AI Cloud Engineering, Enterprise Architecture, Cybersecurity, Legal, Privacy, Governance, and line-of-business leadership.
- Stay current on emerging AI technologies and translate external trends into practical, governed, enterprise-ready recommendations for PG&E.
Technical Stack and Capabilities- Core programming: Advanced Python, reusable package development, APIs, testing, CI/CD, and software engineering practices for AI systems.
- Cloud AI platforms: AWS Bedrock, AWS cloud services, AWS AgentCore/agent runtime capabilities, and integration with enterprise-managed AI gateways and platforms.
- Agentic AI: Agent orchestration, tool calling, memory, multi-step workflows, human-in-the-loop design, multi-agent patterns, and frameworks such as LangGraph, LangChain, or similar technologies.
- LLMOps and MLOps: Model and agent deployment pipelines, observability, monitoring, evaluation, versioning, regression testing, quality gates, and lifecycle management.
- Enterprise integration: Secure integration with enterprise data and systems through APIs, data pipelines, knowledge bases, retrieval systems, vector databases, and governed access patterns.
- Responsible AI: AI risk controls, privacy-aware design, security guardrails, bias and fairness considerations, auditability, model cards, and human oversight practices.
- Data science and ML: Machine learning, statistical modeling, optimization, experimental design, evaluation methodology, and defensible analytical decision-making.
QualificationsMinimum:
- Bachelor's degree in data science, Machine Learning, Computer Science, Engineering, Mathematics, Statistics, Applied Sciences, Physics, Econometrics, Economics, or a related quantitative field, or equivalent work experience.
- 8 years of experience in data science, machine learning, AI engineering, advanced analytics, or related technical roles
- Or 2 years of experience if possess Doctoral Degree or higher, as described above.
Desired:- Experience designing or deploying agentic AI systems, including tool-using agents, workflow automation agents, RAG-based assistants, or multi-agent systems.
- Experience with AWS Bedrock, AWS cloud architecture, enterprise LLM gateways, LangGraph, LangChain, CrewAI, vector databases, knowledge bases, or similar AI engineering technologies.
- Experience with AI evaluation methods, including golden datasets, LLM-as-judge approaches, regression testing, red teaming, performance benchmarking, and production monitoring.
- Experience in a regulated industry such as utilities, energy, infrastructure, financial services, healthcare, or public sector environments.
- Experience working with cybersecurity, privacy, legal, compliance, architecture, and governance stakeholders to deploy technology safely in an enterprise environment.
- Experience contributing to external AI, data science, or machine learning communities through publications, conference presentations, open-source contributions, patents, standards development, or professional leadership.
CompensationPG&E provides the salary range that the company in good faith believes it might pay for this position at the time of the job posting. This compensation range is specific to the locality of the job. The actual salary paid to an individual will be based on multiple factors, including, but not limited to, specific skills, education, licenses or certifications, experience, market value, geographic location, and internal equity. Although we estimate the successful candidate hired into this role will be placed between the entry point and the middle of the range, the decision will be made on a case-by-case basis related to these factors. This job is also eligible to participate in PG&E's discretionary incentive compensation programs.
A reasonable salary range is:
- Bay Area Minimum: $163,000.00
- Bay Area Midpoint: $220,000.00
- Bay Area Maximum: $277,000.00
