Senior AI Infrastructure Engineer

Role Overview

Join VideoAmp as a Senior AI Infrastructure Engineer (individual contributor) to design and own agentic workflow systems that enhance platform APIs and AI-powered customer experiences.

Key Missions

• Architect and own production-critical systems serving live customers.
• Build multi-tenant capabilities including:
  • multi-tenant isolation
  • rate limiting
  • caller attribution as enterprise customers use MCP tools
• Drive evaluation-driven development by:
  • defining golden scenario suites
  • creating automated CI/CD evaluation pipelines
  • implementing regression detection across workflows

Responsibilities

• Design and implement systems that support production agent loops with context/state management across turns.
• Improve performance and cost via prefix caching and coordinated multi-step execution.
• Produce design docs, postmortems, and system diagrams to align technical and non-technical stakeholders.
• Own production API design and operations, including on-call and incident response.
• Collaborate with internal engineering, product, and operations teams to align and roll out shared infrastructure.

Requirements

• Experience designing agent loops for production, including:
  • context and state management across turns
  • prefix caching for latency/cost
  • multi-step task execution coordination
• Experience building automated test/evaluation suites for LLM systems:
  • golden scenarios
  • regression detection
  • quality gates in CI/CD
• Experience with MCP or equivalent tool-layer abstractions for exposing platform capabilities to agents (and/or agents consuming them).
• Strong background in resource-based API design and operating developer-facing platform APIs at scale (including on-call/incident response).
• Deep experience with LLM APIs such as Anthropic or OpenAI (prompting, tool use/function calling).
• Multi-tenant system design experience: rate limiting, caller attribution, tenant isolation, and external support surfaces.
• Working experience across engineering, product, and operations to drive alignment and rollout.
• Familiarity with A2A protocols and long-running durable workflows.
• 6+ years software engineering experience, including 1+ years in AI/ML infrastructure, LLM platform engineering, or agentic systems.
• Go (Golang) is required (backend is Go-first). Python and SQL proficiency also expected.

Nice-to-haves

• AdTech/media measurement familiarity (linear TV, OTT, CTV, digital video).
• Open-weights model experience (e.g., Llama, Mistral).
• Familiarity with Rust.
• Fine-tuning experience (e.g., LoRA/PEFT).