Senior AI Engineer, Media

Role

ACQ Media is the in-house media engine behind , producing long‑form series, podcasts, and daily social content at studio scale for millions of viewers. We are building an AI‑native production system that makes our team measurably faster, more consistent, and more capable than any team operating without it.

The Senior AI Engineer, Media is the technical builder who ships that system. You design, construct, and maintain the agentic workflows, internal AI tools, and generative media pipelines that power every department in our studio operation. This is a production engineering role: prototype rapidly, but ship to a bar production teams can rely on every day.

This is a senior individual‑contributor engineering role. You are not managing a team, and you do not own the strategic roadmap — the Head of AI does. You make architectural and shipping decisions inside that roadmap, flag what will not work, and ship on a timeline production can plan against.

You report to the Head of AI, Media with a dotted line to the CTO. A second hire — a generative‑media specialist focused on the creative tooling layer — will follow this one and partner with you on the deepest creative work.

• Design, build, and maintain agentic AI workflows, owning the full lifecycle from prototype through production monitoring. Framework choice is yours — Lang Chain/Lang Graph, CrewAI, Google ADK, Anthropic’s managed agents, OpenAI Agents SDK, or no framework at all if that is the right call

• Build and operate generative media pipelines (video, audio, captioning, thumbnails, long‑form to short‑form repurposing) using Comfy
  
  UI, Fal, and direct model APIs. You will partner with the generative‑media specialist on the deepest creative‑tooling work once that role is filled

• Ship internal AI tools and dashboards (including search, retrieval, and MCP‑exposed surfaces) that give Post, Art, Content, and Marketing teammates direct access to AI workflows without engineering support

• Build evaluation, monitoring, and failure‑recovery systems for every workflow you deploy, including managed compute, model hosting, and API cost optimization

• Partner with the CTO and Tech team on infrastructure, deployment, security, and IP and data handling standards across all AI systems

• Translate department pain points into engineered solutions with measurable outcomes, train the receiving teams, and document every system you ship

• 5+ years building production software, with strong Python, API design, databases, and cloud infrastructure fundamentals. Enough frontend competence to ship a usable internal tool without a dedicated frontend engineer

• 2+ years building production agentic AI systems — framework‑based (Lang Chain, Lang Graph, CrewAI, Google ADK, Anthropic managed agents, OpenAI Agents SDK, Auto Gen, or equivalent) or framework‑free with custom orchestration. Hands‑on experience with RAG (including multi‑modal), evaluation, and prompt engineering

• Strong instincts for agent‑tool interface design. Direct MCP experience is a plus; the bar is the ability to reason clearly about tool surfaces, schemas, granularity, and statefulness

• Proven track record building internal tools, dashboards, and API integrations that non‑technical users actually adopt. Comfort partnering directly with creative roles — editors, writers, producers, designers — and translating their workflows into systems

• Demonstrated ability to operate in ambiguity, prototype rapidly, and ship production‑grade systems without a fully scoped specification

• Direct, professional experience running media production AI workflows with tools like Comfy
  
  UI, Fal, and similar generative platforms (including video diffusion, LLMs, and audio generation models)

• Experience working inside or directly supporting media, video production, or content operations teams

• Two or more production‑grade AI workflows deployed and actively used by department teams within 90 days, with monitoring and evaluation in place. The first will be a multi‑modal search and retrieval tool for our long‑form content library, surfaced both as a web portal and an MCP server

• Measurable reduction in…