Senior AI Platform Engineer

Alpaca is building an AI Enablement function to move the company from scattered AI experimentation to durable, company-wide productivity gains. As Senior AI Platform Engineer, you will own the technical capability layer that makes that possible.

You will build and maintain the platform infrastructure, connectors, execution patterns, and self-service tooling that lets engineering and business teams use AI safely at scale by turning one-off setups and ad hoc tooling into reusable infrastructure, secure deployment standards, and productized onboarding.

This role requires hands-on experience building with agentic AI systems. This is not classical ML pipelines, but the emerging stack of LLM-powered agents, tool use, evaluation loops, and autonomous workflow execution. You should know this space well enough to make opinionated decisions about execution patterns, isolation boundaries, and what "safe and reliable" actually means for agents with real tool access.

This is a startup-inside-a-scaleup role. You will need to move fast with limited resources while meeting the standards a regulated fintech environment requires. You will work closely with DevOps, Security, IT, and engineering teams across the company.

• Own the connector and service integration layer that powers AI workflows across the company.
• Design and ship execution environments for agents and higher-autonomy AI workflows, including isolation boundaries and access controls.
• Build reusable platform services, golden paths, and self-service templates that reduce setup friction for teams building on AI.
• Productize onboarding so it works reliably for both developers and non-developers without depending on manual intervention or tribal knowledge.
• Define and enforce technical standards for agent execution, evaluation loops, and deployment.
• Partner with Security and IT to ship deployable patterns for higher-risk AI capabilities.
• Own the AI governance layer: access controls, audit trails, approval criteria, and deployment boundaries for agentic workflows.
• Set the reliability, observability, and operational bar for AI-specific infrastructure.
• Act as the technical escalation point when onboarding or platform issues block rollout.
• Reduce the company's dependence on individual heroics by turning exception handling into repeatable paths.

• 8+ years in software, platform, infrastructure, or adjacent engineering roles.
• Hands-on experience building agentic AI systems: LLM-powered workflows, tool-calling agents, evaluation loops, or autonomous execution — using frameworks like the Claude SDK, Google Agent Development Kit (ADK), LangGraph, or similar. Not classical ML or data pipelines.
• Direct experience with GCP. We run on Google Cloud and you should be comfortable there.
• Strong experience with APIs, auth, OAuth, secrets, CLI tooling, and deployment patterns.
• Cloud-native systems experience with containers, orchestration (Kubernetes), and infrastructure-as-code.
• Experience implementing AI governance controls: access boundaries, audit logging, approval workflows, and safe deployment standards for higher-autonomy systems.
• Comfortable operating in both fast-moving, low-process environments and more structured, compliance-aware ones. You know when to move fast and when to slow down.
• Strong bias toward simplification, standardization, and operational reliability over clever one-off solutions.
• Excellent communication skills with the ability to work across engineering, security, and non-technical stakeholders.

• Shipped production agentic systems with real external tool access (filesystems, APIs, staging systems).
• Hands-on experience with Google Agent Engine (Vertex AI Agent Builder) or equivalent managed agent execution platforms.
• Direct experience with AI-native coding environments (eg. Cursor, Claude Code). You have strong opinions on how these tools fit into a real engineering workflow.
• Designed or operated agent sandboxing, isolation, or evaluation frameworks.
• Built self-service developer platforms or golden paths used by multiple teams.
• Has startup experience and knows how to build durably with limited resources.
• Familiarity with fintech, regulated environments, or compliance-aware deployment.