Charlie Yang
Software Development Engineer at AWS Kiro
I work on agentic AI developer experiences and context systems, with an additional research focus on medical imaging and clinical AI.
Email · Google Scholar · GitHub · LinkedIn
About
I’m a software engineer at AWS Kiro, where I work on agentic AI developer experiences.
Alongside product engineering, I contribute to medical imaging and AI agent research at the University of Colorado Anschutz School of Medicine. I enjoy turning ambiguous research problems into dependable products—balancing technical depth with usability, performance, and clean execution.
Highlights
- At AWS Kiro, designing an AI automation framework for agent-driven workflows—standardizing multi-step tool calling and orchestration (MCP), execution state management, and failure recovery for reliable long-running agent sessions.
- Evolving LLM context and memory systems (compaction, retrieval, context scheduling under token budgets), reducing inference cost and latency while preserving response quality.
- As an intern, designed and shipped Kiro IDE's core distributed context compaction system—hierarchical compression with RAG on Amazon S3 Vectors and OpenSearch Serverless, achieving 80–90% compression—and drove its production launch serving 100K+ users.
- Contributor to KiroCrew, an open-source multi-agent workspace built on kiro-cli and the Agent Client Protocol (ACP): session-resilience, redaction, and sandboxing improvements.
- Research on reliable AI for medical imaging: reliability stress tests and decision-time routing for chest X-ray vision–language models (CHASE 2026) and confidence-gated cloud–edge cascade triage (Smart Health, 2026), plus a U.S. provisional patent on an intelligent radiology reporting platform.
- Earlier, built a real-time ICU patient monitoring system at the Chinese Academy of Sciences, Institute of Automation, securing a software patent and a national grant.
Experience
-
Software Development Engineer, Amazon Web Services · Kiro
Building agentic AI developer experiences and evolving the context systems that make long-running workflows more capable, reliable, and intuitive.
-
Research Assistant, University of Colorado Anschutz School of Medicine
Researching medical imaging and AI agents in a remote, part-time role.
-
Software Engineer Intern, Amazon Web Services · Kiro
Led the design and productization of a modular context management system combining hierarchical compaction, memory, and retrieval.
-
Research Intern, Brown University Health
Developed an AI-powered medical imaging platform for report generation, image analysis, and clinical comparison workflows.
Earlier experience
- · Personal Trainer, Brown University Athletics · View details
- · Lead Web Developer & Designer, Rhode Island Coalition for Israel · View details
- · Algorithm Engineering Intern, Shukun Technology · View details
- · Software & ML Engineer Intern, Chinese Academy of Sciences · View details
- · Research Intern, University of Science and Technology Beijing · View details
- · Research Intern, Nanyang Technological University · View details
Academic
Research interests: Multimodal LLMs · Medical Image Analysis · AI in Healthcare · Agentic AI
Citations: 47 ·
h-index: 2 ·
i10-index: 1 ·
View Google Scholar
Updated September 2026
Publications
Metrics and publication citations are synchronized from Google Scholar.
- Reliability Stress Tests and Decision-Time Routing for Chest X-ray Vision-Language Models 2026 IEEE/ACM Conference on Connected Health: Applications, Systems and Technologies, 2026.
- Confidence-gated cloud-edge cascade triage via variational risk minimization for medical imaging Smart Health · 100689, 2026.
- A Unified Platform for Radiology Report Generation and Clinician-Centered AI Evaluation medRxiv · 2025.07.07.25331018, 2025. cited by 1
- The AI Challenge: A Turing Test Pilot Study of Attendings and Residents in Identifying AI-Generated Content Meta-Radiology · 100199, 2025.
- Understanding the brain with attention: A survey of transformers in brain sciences Brain-X · 1(3), e29, 2023. cited by 44
- Optimization of CNN for Diagnosis on Lung Disease by Lung Segmentation and Rib Suppression 15th International Symposium on Computational Intelligence and Design, 2022. cited by 2
Education
-
Master of Science in Computer Science, Brown University
Focus on AI, machine learning, and software systems.
-
Bachelor of Engineering in Artificial Intelligence, University of Science and Technology Beijing
Graduated with honors · GPA 3.83/4.0.
Skills
- Languages: TypeScript · JavaScript · Python · Java · C/C++ · SQL · Go
- AI & Data: LLM Systems · RAG · MCP · PyTorch · TensorFlow · Computer Vision
- Frameworks: React · Node.js · Flask · Spring Boot · MySQL
- Tools: AWS · Docker · Git · Google Cloud · Vite · Figma
Contact
I’m always glad to connect with people working on thoughtful AI products, developer tools, and healthcare technology.
charlieyangxinye@gmail.com · LinkedIn · Google Scholar · GitHub