Principal Research Programmer (Epic Developer Assistant)
- Vancouver, Canada
- Lead/Staff
- Posted
Skills
Required
- programming
- Unreal Engine
- C++
- Python / web backend language
- game engine
- tooling / editor
- scripting
- cloud-native
- LLMs
- software programming
- engine-side
Preferred
- d'Unreal Engine
Join Epic Games as a Principal Research Programmer, where you'll integrate AI into Unreal Engine. You'll design and ship systems that allow LLMs to operate within the engine, working across the full stack from engine tooling to cloud backend. You'll also evaluate and integrate emerging LLM capabilities and define the technical direction for LLM-driven engine automation. Strong programming skills in C++ and Python, experience with cloud-native services, and a background in game engine architecture are required.
Responsibilities
- Concevoir et expédier l'ensemble des systèmes permettant aux LLM d'agir à l'intérieur du moteur, en intégrant l'architecture du moteur de jeu, les services cloud et le fonctionnement des LLM.
- Concevoir et construire des agents et des outils pilotés par LLM qui fonctionnent directement à l'intérieur d'Unreal Engine, y compris les outils C++/Python côté moteur et le scripting de l'éditeur.
Requirements
- Évaluer et intégrer les capacités émergentes des LLM (utilisation d'outils, orchestration multi-agents, gestion du contexte) dans le moteur d'Epic et les outils de pipeline.
- Excellent written and verbal communication skills
BS/MS in Computer Science or a related field
- Strong software programming background with hands-on experience in a game engine (Unreal Engine strongly preferred), including engine architecture and tooling/editor scripting
- Excellent programming skills across C++ (engine-side) and Python and/or a web backend language
- Experience with cloud-native services and infrastructure: web APIs, databases, and service architecture and deployment
- Ability to work full-stack and reason about tradeoffs across engine code, backend services, and a web front end
- Practical experience building or shipping products powered by LLMs (agents, tool-calling, RAG, prompt/context programming), or a strong conceptual understanding of how LLMs work
jobapplier