Software Engineer

What You Need to Bring - Must 

•     Agentic AI 

◦     You understand how agents actually work: LLMs, tool calling, orchestration, memory, RAG 

◦     You can explain how an agentic system is designed from first principles, where it breaks, and why

◦     You've been actively building and shipping with AI over the past year and you can walk us through what you learned

•     Full-stack engineering 

◦     Can architect and build end-to-end systems: frontend, backend, database and knows the right patterns to reach for

◦     Understands what's happening under the hood, not just how to call the abstraction

◦     Cloud fluent: can deploy on AWS/GCP/Azure, comfortable with pipelines, containers, IaC, and IAM; enough to not be blocked

◦     Uses modern AI tooling (Cursor, Copilot, etc.) to ship faster and can articulate how 

Domain Knowledge 

We work across a wide range of media & entertainment business contexts. The strongest candidates come in with serious depth in at least one domain beyond engineering.  

•     Data Science / BI / Analytics

•     Machine Learning & Deep Learning

•     Computer Vision

•     Data Engineering or Distributed Computing

•     Cloud Engineering

•     Marketing, Sales, or Business Operations

•     Cognitive Science / Linguistics / NLP

Traits That Matter as Much as Skills

•     Business curious: You ask why the problem exists, not just how to build the solution. You push back when the ask doesn't make sense.

•     Adaptable: The market is moving too fast for rigidity. If you haven't updated your mental model of AI in the last 6 months.

•     Thought partner:  You sit with the business, iterate, communicate, and elevate the conversation.

•     Genuinely loves to build:  Not the idea of it, the actual work. Hands on keys, something ships, you learned something, you do it again.

•     Has the itch:  Intrinsic drive that doesn't require external motivation. You were probably thinking about this stuff last night.

•     Thinks under the hood:  Doesn't need to know every implementation detail but understands basic theory well enough to reason about what's breaking and why.

•     Builds without instructions:  Ambiguity is fuel, not friction. You figure it out, document what you learned, make it repeatable.