Internal AI Platform Engineer

This is a hands-on role for someone who can take messy real-world workflows and turn them into reliable internal systems.

You'll work closely with leadership, domain owners, and a knowledge engineer focused on workflow design, process mapping, and evaluation. The job is to understand how work actually happens, identify where AI creates leverage, and build the systems that make it dependable in practice.

You'll help build:

• Orchestration and control-plane infrastructure for AI across functions
• Connectors across tools, workflows, documents, and operational signals
• Permissions, tracing, context services, and deployment plumbing
• Sandboxed environments for testing AI systems against real operational data
• Internal tools that bridge the gap between a prototype and something a team can rely on every day

This is not a generic AI engineer, prompt engineer, model-training, or applied-LLM feature role.

Requirements

What we're looking for:

• Senior, staff, principal, or architect-level engineering experience
• Strong systems depth across backend services, integrations, data flows, and application logic
• Evidence of building internal platforms or workflow systems, not just end-user AI features
• Practical daily use of LLMs as building tools
• Clear judgment about deterministic versus model-driven system boundaries
• Experience with human review, escalation, auditability, and operational safeguards
• Willingness to work closely with teams to understand real workflows, not just stated requirements

You may be a strong fit if:

• You've built internal platforms, shared services, or workflow systems used by multiple teams
• You've designed connectors, integration patterns, tool contracts, or context layers
• You've built AI systems for enterprise operations, security-sensitive workflows, compliance-heavy domains, or internal business processes

You are probably not a fit if:

• Your background is mainly generic AI/ML feature work
• Your strongest work is consumer search, ranking, recommendation, or shopping systems
• Your work is mostly model training or research without substantial systems ownership
• Your background is mostly prompt engineering, coursework, or hackathon projects