AI/ML Engineering and Architecture Manager

Role Synopsis

bpx energy is building an enterprise AI capability that can scale safely and deliver real operational value. We are seeking a AI/ML Engineering and Architecture Manager to own how AI is engineered, deployed, operated, and governed across the company.  This is a deeply technical AI engineering leadership role, not a traditional software or application engineering position. Success requires hands‑on experience operating production AI/ML systems at scale, including ModelOps/MLOps, platform engineering, and runtime reliability. 

For 2026, the role will lead centralized AI/ML engineering with federated intake and enablement, establishing the standards, platforms, and “paved roads” required to move fast without creating risk. As these foundations mature, the role will define and enable a clear pathway to federated AI/ML engineering for domain teams (targeted for 2027, based on readiness). 

This role shapes AI capabilities that support enterprise‑critical operations, system‑level decisioning, and scalable automation across bpx energy, where engineering rigor and operational reliability have material business impact. “Federated” refers to the architectural and operating model design, enabling future distributed delivery once enterprise standards and governance are in place; engineering execution remains centralized in 2026. 

What You’ll Do 

• Define and own bpx's enterprise AI/ML engineering architecture, standards, and operating model 

• Operate and evolve modern AI/ML platforms (e.g., Databricks or equivalent) for production‑grade use 

• Own ModelOps/MLOps, including deployment automation, monitoring, drift detection, evaluation, and lifecycle management 

• Create the paved road for AI delivery: reusable patterns, approved integrations, tiered access controls, and observable operations 

• Embed security, governance, and audit-ability directly into engineering workflows 

• Enable domain teams through federated intake and enablement, while preventing uncontrolled “wild west” development 

• Partner closely with data, BI, security, and architecture leaders to enforce clear platform roles and interoperability standards 

• Build and lead a small, high‑impact platform and enablement team 

• Own enterprise AI/ML platform engineering and technical enablement, including Databricks as the primary AI/ML engineering platform, and technical oversight of Palantir’s integration with AI/ML services to ensure fit‑for‑purpose execution, platform discipline, and adherence to enterprise standards. 

• Palantir value delivery, workflow ownership, and ontology ownership remain outside this role; this role owns the technical patterns, AI integration standards, and guardrails that determine how and when Palantir consumes AI outputs. 

What This Role Is Not 

• Not a traditional software engineering or application platform leadership role 

• Not focused on building individual AI use cases or owning business value delivery 

• Not a “hands‑off” architecture governance position 

Required Qualifications

• 5 years of demonstrated experience operating production AI/ML systems in large or sophisticated environments 

• Prior, multi-year ownership of ModelOps/MLOps, including deployment, monitoring, evaluation, retraining, and retirement 

• 5 years hands‑on experience with modern AI/ML platforms (e.g., Databricks, Snowflake, Palantir or equivalent), beyond experimentation 

• Proven track record to design and operate AI engineering systems that differ materially from traditional application or data platforms 

• Strong fundamentals in security, controls, and operational reliability (IAM/RBAC, audit logging, incident management) 

• Candidates whose experience is limited to traditional software engineering, application platforms, or data engineering—without direct responsibility for AI/ML operations at scale—will not be considered. 

Why This Role Matters 

This role establishes the engineering system that makes enterprise AI possible — enabling speed with guardrails, reducing duplication and risk, and ensuring bpx energy can scale AI responsibly over time.