LLM Engineer

Supercell AI team drives the companywide AI agenda and acts as the subject matter expert team for all AI related topics. Supercell is known for its small teams that can deliver global hit games. We want to equip our employees with the right tools to keep them super efficient and effective in the AI era, and automate internal and player facing repetitive tasks using agents and bots. We are now preparing for a future where humans and agent team members works seamlessly side-by-side.

We are looking for an experienced LLM Engineer that has built sophisticated production grade agentic LLM applications. Do you always know which are the strengths and weaknesses of the latest LLM models? Are you a prompt engineering, RAG and MCP wizard? Do you know how to build smart adaptable agentic applications? If yes, we can provide you with an interesting role in a high-impact and collaborative team.

In this role you will act as our LLM and agent expert working closely with our teams implementing LLMs into our workflows. You might spend some time for example embedded in one of our amazing game teams, build bots for serving our players, build agents for drawing insights from social media, help our experts automate parts of their work or design completely new virtual team members for us. 

Responsibilities 

• Monitor developments in the LLM and agents space and assess suitability of relevant technologies for our use

• Maintain a suite of recommended LLM and agent technologies and document the best ways how our teams can access these technologies safely

• Work closely with our game and other teams in implementing LLM and agentic projects helping our employees be more productive and providing a better experience for our players 

• Support our internal teams in applying LLM and agentic technologies to streamline their work and provide expert advice on best practices

• Define and implement a robust evaluation and monitoring systems for our production LLM powered systems ensuring safety, reliability and performance at scale

Requirements

• Experience building production grade LLM and agentic applications

• Expert level knowledge of core LLM and agentic concepts including RAG, CoT, reasoning models, memory, fine-tuning, tool use, self-correction

• Good understanding of the different multi-agent system design patterns and their suitability for different kinds of use cases

• Good software development skills and ability to quickly build robust production ready solutions for internal use cases

• Good problem solving skills

• Ability to communicate about technical concepts clearly to non-technical audiences

• Ability to work independently and as part of a team

• Good verbal and written communication skills in English

• Positive can do attitude

• Experience from gaming industry or passion for gaming

• PhD in ML 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