Software Development Engineer, Specialist AI Tooling, Specialist Technology Team

AWS Specialist Technology Team (STT) is seeking a Software Development Engineer to join the Specialist AI Tooling engineering team - the group building AI-native, agentic solutions that are fundamentally transforming how thousands of AWS technical specialists and field team members develop expertise, get matched to customer engagements, and how AWS service teams consume customer feedback at scale. If you want to build production agentic AI systems that impact thousands of users daily - not research prototypes, but real AI agents solving real problems - this is the role for you. You will design, build, and operate core components across four interconnected products that collectively power the intelligence layer of AWS's specialist ecosystem: - AWS Learning Agent - A new generative AI-powered, agentic learning system that replaces static training paths with personalized, cross-domain skill development for every builder in the AWS technical field. Powered by a structured skill graph, persistent learner profiles, and the Specialist Knowledge Vault, the Learning Agent delivers adaptive learning journeys via a custom console, in Kiro, Claude, Quick and wherever builders work. - Intelligent Routing Agent - A new AI-driven routing engine that matches 136K+ annual specialist requests to verified domain experts using validated skill profiles rather than manually maintained lists - reducing resolution time (currently 7.72 days) and enabling precision matching at scale. - Technical Field Community (TFC) Hub - The platform powering 46 Technical Field Communities, managing member engagement, competency tracking, learning path orchestration, and community-driven content curation across AWS's specialist network. - Service 360 - A GenAI summarization and prioritization engine that aggregates customer feedback signals (PFRs, SpecReqs, SFDC Opportunity data) into categorized, prioritized service recommendations - automating what was previously a manual closed-loop feedback process to influence the roadmap for AWS public service teams. This is a high-impact, AI-first engineering role where you will build agentic systems, ship production GenAI features, and solve hard problems at the intersection of knowledge graphs, multi-agent orchestration, LLM-powered personalization, and retrieval-augmented generation - all at AWS scale, using the very AI services AWS builds for its customers. Key job responsibilities - Design and implement production agentic AI systems using LLMs, knowledge graphs, RAG, and multi-agent orchestration - delivering personalized learning paths and intelligent expert-matching to thousands of field members daily. - Build and optimize GenAI-powered pipelines including summarization, content ranking, skill assessment, and intelligent routing - ensuring low latency, high quality, and cost efficiency at scale. - Develop and iterate on prompt engineering strategies, evaluation frameworks, and guardrails for production LLM systems - ensuring quality, safety, and reliability. - Implement embedding and retrieval strategies for a knowledge base optimizing how domain expertise is indexed, retrieved, and contextualized for agent consumption. - Design and build scalable distributed systems and APIs that power the TFC Hub platform, serving thousands of community members across 46 technical domains. - Drive operational excellence for production services - owning CI/CD, monitoring, alarming, on-call, and incident response for systems serving thousands of daily users. - Contribute to technical strategy and architecture decisions, proactively identifying opportunities to improve system performance, reliability, and developer velocity. - Collaborate with Product Management, Science, and field stakeholders to translate ambiguous customer problems into well-scoped technical solutions.

You will be joining a high-growth engineering organization that is at the forefront of applying generative AI and agentic technologies to transform how