IT Engineer, Internal AI Infrastructure

The shape of it

In the posting’s own words

We're a small, high-leverage team organized around three pillars: Platform & Tooling, Infrastructure & Hosting, and Applied AI Builds. This role anchors the Infrastructure & Hosting pillar. This is the runtime layer everything else depends on.

What it asks for · 5

• 5+ years of software engineering experience, with significant time in infrastructure, platform, or backend engineering
• A track record of building internal platforms or paved paths that other engineers and teams adopted, not just maintained, with excellent communication skills collaborating across security, IT, and business teams
• Strong system design skills and experience architecting resilient, observable production systems
• Hands-on experience with LLM operations: model gateways or routing, agent runtimes, evals, or cost/usage instrumentation
• Experience with identity and access management in an enterprise environment (e.g., Okta, SSO, secure service-to-service auth)

Also a plus

• Experience building internal AI platforms or workspaces (the Ramp Glass / JPMorgan LLM Suite category) or developer platforms at a comparable scale
• Familiarity with MCP, agent orchestration frameworks, or skill/plugin registry patterns
• Experience working within audit and compliance programs (SOX/ITGC) or building systems with audit requirements
• A history of being the first infrastructure hire on a small team, comfortable setting standards from scratch across a broad scope of responsibilities

What the job covers

• Design and build the hosting golden path for internal AI applications, bots, and agents, including runtime patterns for long-running workflows, scheduled jobs, and agent execution
• Build the model routing layer that directs tasks across frontier and lower-cost models based on cost, capability, and policy with a full audit trail
• Stand up observability, usage analytics, audit logging, and per-team cost attribution for AI workloads across the company
• Integrate identity, access, and security patterns (Okta-connected by default) so every internal AI app is safe on day one, including the sanctioned path for workflows touching customer, financial, or people data
• Partner with engineers on our product AI infrastructure teams to share gateways and telemetry rather than building parallel systems
• Help shape the team's platform strategy, roadmap, and engineering best practices as a founding member of the infrastructure pillar

Tools and skills named

Words the posting leans on

• infrastructure8×
• platform8×
• experience7×
• design6×
• internal6×
• agent5×
• audit5×
• build5×
• building5×
• model5×
• engineering4×
• hosting4×
• layer4×
• path4×
• patterns4×
• security4×

Counted from the posting after the mission statement and the legal notices are set aside. The ones near the top are the ones a screener is looking for.

More open at Figma

• Account Executive, Emerging Enterprise (Berlin, Germany)Berlin, Germany
• Account Executive, EnterpriseSan Francisco, CA • New York, NY • United States
• Account Executive, Enterprise (Bengaluru, India)Bengaluru, India
• Account Executive, Enterprise (Berlin, Germany)Berlin, Germany
• Account Executive, Enterprise, Mandarin Speaking (Singapore)Singapore
• Account Executive, Enterprise (Paris, France)Paris, France