About the Team
At Roku, the Device Monetization and Themed Experiences (DMTE) team, a part of the Advertising organization, is where the magic behind the scenes happens. We’re the engineers who build the high-performance graphics and UI foundations that power the FRAME engine — the core technology enabling Roku’s smooth, responsive, and visually rich experiences on millions of devices worldwide. Our work lives in the middleware and UI layers, and its impact is felt every time a user navigates a menu, enjoys an animated ad, or is wowed by the bespoke experience built to market the latest TV show or Movies. We thrive on solving the toughest technical challenges — from optimizing UI rendering to fine-tuning memory usage — using BrightScript (a JavaScript-like language) to deliver rock-solid, high-performance solutions. We work in a collaborative, high-trust environment where great ideas can come from anywhere, partnering closely with firmware, UI, and product teams to ensure our platform is lightning fast, reliable, and ready for the next generation of streaming innovation. If you love building the invisible foundations that make great user experiences possible, this is the team for you.
About the Role
As a Senior Agentic AI Engineer in the DMTE team, you’ll be the driving force behind embedding intelligent, autonomous AI capabilities into Roku’s platform — reimagining how our systems create, optimize, and deliver experiences to tens of millions of devices. Rather than building UI features directly, you’ll design and deploy agentic AI systems that reason, plan, and act across our development and delivery pipelines — from automating the generation of themed experiences to enabling self-optimizing rendering workflows within the FRAME engine. You’ll leverage your years of experience in software engineering and building world class UI to build AI agents that understand the constraints of embedded, performance‑critical systems and operate effectively within them. This is a role for engineers who think in terms of autonomous systems, multi‑step reasoning, and tool‑use architectures — and who want to fundamentally reshape how a streaming platform is built, maintained, and evolved.
What You’ll Be Doing
- Design, build, and deploy agentic AI systems that autonomously generate, test, and optimize BrightScript/TypeScript code and UI experiences for the FRAME engine
- Architect multi‑agent workflows (planning, tool‑use, retrieval, execution, validation) that accelerate development and reduce manual engineering effort across the team
- Develop AI‑powered automation that understands the constraints of embedded C++ environments and produces performance‑safe, memory‑efficient outputs
- Build and maintain evaluation frameworks to measure agent reliability, accuracy, and performance in production‑critical contexts
- Collaborate with cross‑functional teams (firmware, UI, product) to identify high‑impact opportunities for AI‑driven automation and intelligent tooling
- Stay at the forefront of agentic AI research — including LLM orchestration, RAG, function calling, and autonomous code generation — and bring those advances into our platform
- Mentor engineers across the team on AI‑first development practices and help establish best practices for human‑agent collaboration
We’re Excited if you Have
- Proven experience building agentic AI systems that generate, optimize, and deliver scalable applications in production environments
- Strong software engineering foundation with a track record of shipping reliable, performant systems at scale
- Deep familiarity with modern LLM orchestration tools, prompt engineering techniques, and AI/ML frameworks
- Experience with retrieval‑augmented generation (RAG), function/tool calling, and evaluation‑driven development for AI agents
- A builder’s mentality — you don’t just use AI tools, you create autonomous systems that multiply the impact of entire teams
- Proficiency in TypeScript (or similar languages) and Python for agent development, scripting, and pipeline automation
- A software engineer’s instinct for reliability, testability, and clean architecture — applied to the unique challenges of non‑deterministic AI systems
- Excellent technical communication, root cause analysis abilities, and a proactive mindset with passion for fast iteration
#J-18808-Ljbffr…