SupercellPrincipal LLM Engineer
任职要求
• Deep expertise in building production-grade LLM and agentic applications — including real-world experience with evaluation, monitoring, and scaling. • Strong background in LLM applications, with hands-on experience in AI development tools (e.g. code generation, debugging agents, tool-assisted workflows) and a perspective on how they fit into the broader LLM ecosystem. • Expert-level knowledge of core LLM concepts: RAG, CoT, reasoning models, memory, fine-tuning, tool use, self-correction. • Strong software engineering skills with the ability to deliver robust, production-ready solutions quickly. • A craft mindset: ability to define standards, set direction, and raise the bar across a company. • Excellent communication and evangelism skills — able to sell ideas, influence stakeholders, and build excitement around AI capabilities. • Comfort working independently as well as in small, fast-moving teams. • Positive, pragmatic, and collaborative mindset. • Passion for gaming or experience in the industry a plus. • PhD in ML or equivalent depth of expertise preferred. Relevant technologies and platforms • LLMs: GPT, Claude, Llama, Mistral, Gemma • Agent technologies: OpenAI Agents & GPTs, Anthropic Agents, MCP, LangGraph, Amazon Bedrock Agents • ML: PyTorch, TensorFlow, ONNX • Data: Python, Databricks, Spark • Cloud: AWS, GCP • Infra: Docker. Kubernetes, Redis • Vector databases: Pinecone, Chroma, pgvector
工作职责
• Act as the craft owner for LLM engineering at Supercell — setting direction, sharing best practices, and raising the bar for what “great” looks like in LLM-powered systems. • Drive adoption of a wide range of LLM and agentic applications (e.g. in-game bots, player support, social insights, internal productivity tools). • Champion AI development tools as a particularly high-leverage use case — helping our teams accelerate workflows, code smarter, and build better systems faster. • Partner closely with game and functional teams to evangelize LLM capabilities and turn vague opportunities into concrete solutions. • Serve as a bridge between technical and non-technical stakeholders, clearly communicating the strengths, limitations, and business value of different approaches. • Monitor emerging developments in the LLM and agents space; assess technologies for safety, reliability, and performance; and build our internal “stack” of recommended approaches. • Lead the design and implementation of robust evaluation, monitoring, and safety frameworks for production LLM-powered systems. • Rapidly prototype and deliver production-grade systems — but also coach, mentor, and enable other engineers to build confidently themselves.
- Keep up to date with and utilize the latest developments in LLM system optimization.- Take the lead in designing innovative system optimization solutions for internal LLM workloads.- Optimize LLM inference workloads through innovative kernel, algorithm, scheduling, and parallelization technologies.- Continuously develop and maintain internal LLM inference infrastructure.- Discover new LLM system optimization needs and innovations.
• Lead the software development in C/C++, Python, and in GPU languages such as CUDA, ROCm, or Triton• Analyze metrics and identify opportunities based on offline and online testing, develop and deliver robust and scalable solutions.• Work with cutting-edge hardware stacks and a fast-moving software stack to deliver best-of-class inference and optimal cost.• Engage with key partners to understand and implement inference and training optimization for state-of-the-art LLMs and other models.
Architect, build, and optimize secure and performant AI platform services that power Microsoft Copilot and other next-generation AI scenarios. Provide technical leadership across teams to define long-term architectural direction and drive engineering excellence. Collaborate with infrastructure, platform, product, and research teams to design and deliver scalable, production-grade AI services. Write high-quality, well-tested, secure, and maintainable code and promote high standards across the team. Tackle technically ambiguous or cross-boundary problems, remove roadblocks, and drive delivery across multiple teams or organizations. Lead technical design discussions, mentor senior engineers, and foster a strong engineering culture within the team. Embody Microsoft’s Culture and Values, and help shape the direction of the engineering team and broader organization.
Architect, build, and optimize secure and performant AI platform services that power Microsoft Copilot and other next-generation AI scenarios. Provide technical leadership across teams to define long-term architectural direction and drive engineering excellence. Collaborate with infrastructure, platform, product, and research teams to design and deliver scalable, production-grade AI services. Write high-quality, well-tested, secure, and maintainable code and promote high standards across the team. Tackle technically ambiguous or cross-boundary problems, remove roadblocks, and drive delivery across multiple teams or organizations. Lead technical design discussions, mentor senior engineers, and foster a strong engineering culture within the team. Embody Microsoft’s Culture and Values, and help shape the direction of the engineering team and broader organization.