About the role
Crunchyroll is growing and changing, presenting unique challenges and opportunities to support millions of anime fans around the world. The AI/ML team provides seamless help to our internal stakeholders, ensuring an exceptional experience for all Crunchyroll fans.
- Architect and implement end-to-end LLM-based applications, including chatbots, assistants, search agents, and knowledge discovery tools.
- Design and deploy scalable RAG and GraphRAG pipelines using vector databases (e.g., Opensearch, Pinecone, Mongo) and document chunking strategies.
- Fine-tune and optimize foundation models (e.g., Bedrock, Claude, Nova) for domain-specific use cases.
- Own prompt design and evaluation - build prompt libraries and frameworks that enable rapid experimentation and reuse.
- Develop APIs and microservices to serve LLM-based applications with real-time latency and uptime guarantees.
- Collaborate with data, platform, and backend teams to integrate LLM capabilities into broader Crunchyroll product experiences.
- Monitor and evaluate LLM systems in production for hallucination, latency, cost, and feedback loops.
- Stay up to date with the latest in LLM research, tooling (LangChain, LlamaIndex), and ecosystem advancements.
About You
We get excited about candidates, like you, because...
- Bachelor's or Master's degree in Computer Science, Machine Learning, AI, or a related field.
- 5+ years of experience building production ML/AI systems, with 2+ years specifically focused on LLM applications.
- Hands-on experience with production-ready RAG systems using tools like LangChain, LlamaIndex, or custom pipelines.
- Hands-on experience with GraphRAG, LLMs, Graph and Vector Databases.
- Deep understanding of prompt engineering, prompt templating, and prompt evaluation strategies.
- Experience working with major foundation models (OpenAI, Anthropic, Nova, Bedrock etc.).
- Familiarity with building semantic search systems and working with vector DBs at scale.
- Strong backend/API development experience using TypeScript and knowledge of deploying services to cloud environments.
- Python skills and experience working with AWS cloud platform and MLOps tools (e.g., AWS Bedrock, SageMaker, Databricks, Lambda, ECS/Fargate).
About the Team
The R&D team is dedicated to developing, testing, and validating robust and scalable machine learning models that drive business objectives. Our focus includes enhancing operational processes through AI/ML solutions, such as trend analysis, anomaly detection, and the deployment of large language models (LLMs) for tasks like querying system health. Another major focus area is preserving, and improving customer experience and retention. We closely work with our stakeholders to ensure AI/ML objectives are clearly defined.