• Deploy and manage LLM endpoints and AI pipelines using AWS Bedrock, EC2, and SageMaker • Integrate OpenSearch indexes via Bedrock Knowledge Bases • Build and maintain secure internal APIs • Support in configuring and monitoring AI model usage • Work with the Technical Lead to integrate STT/TTS • Partner directly with our Full Stack Engineer • Design and build a flexible, secure, and scalable MLOps foundation

• Build system-level software: our challenges aren't your usual kube-fits-all variety, so be prepared to build things from scratch based on first principles. • Have a strong understanding or curiosity of OS level primitives. • Be a student of both distributed systems and simplicity. • Roll with the boring: build Golang/Rust GRPC services from scratch to expose your complex capabilities to the wider company in a scalable and performant manner. • This is a high impact, high agency role with direct effect on company culture, trajectory, and outcome. • You may be oncall from time to time in this role.

• Architect and deploy Open Source Data OSS (Trino, Spark, Pinot, Druid) on Amazon EKS • Maintain a 100% hands-on approach, utilizing IDEs, Git, and strict coding best practices to manage infrastructure as code • Navigate complex customer hierarchies and technical landscapes to identify, propose, and implement bespoke solutions • Design and implement novel, high-risk solutions that have not been previously established • Act as a subject-matter expert in Open Source technologies, ensuring all builds are well-architected and production-ready

• Design and build scalable platform components leveraging advanced retrieval via query planning, semantic and hybrid search, metadata-aware search, and LLM generation • Design and build optimized indexing pipelines for structured and unstructured data • Build backend services for semantic and hybrid retrieval, knowledge graph construction, and retrieval orchestration • Improve retrieval quality through evaluation and observability frameworks • Design APIs for internal and external user and agentic consumers • Optimize latency, throughput and cost across large-scale inference and retrieval workloads • Drive technical direction for reliability and security