Software Engineer | AI Systems | Real-Time & Data Platforms
MS + BS Computer Science @ Arizona State University

I build AI-powered, data-heavy products that real users depend on spanning full-stack web, distributed systems and agent-based AI. I care deeply about performance, correctness, and user trust especially when AI sits directly inside core workflows.

Previously shipped production systems at Zoom (Next Lab), built NSF-funded education platforms and developed AI products used by hundreds of users.

Software Engineer, Next Lab (Zoom Innovation Lab)

• Built and shipped production-ready prototypes for Zoom’s Innovation Lab used by 5,000+ hybrid learners
• Worked on real-time communication systems using WebRTC, React, Docker, and AWS
• Tuned signaling paths and interaction flows to reduce end-to-end latency in live classroom experiences
• Operated in environments where performance, reliability, and UX were immediately user-visible

Research Aide – Cognition Lab (ASU)

• Ran 1,500+ multi-step agent interactions across clinical simulation benchmarks (MedAgentBench, AgentClinic)
• Built evaluation pipelines and error taxonomies to analyze reasoning drift, hallucinations, retrieval, and factuality failures
• Designed reproducible AI experimentation workflows using Python, Docker, SQL, and AWS
• Focused on improving model reliability and pushing accuracy beyond current baselines (~67%)

Software Engineer – VizCoach (NSF-funded, $300K)

• Engineer on a classroom orchestration and analytics platform used by 150+ students per semester
• Built real-time dashboards, chat, and authoring tools using React, Go, SQL, WebSockets
• Maintained 100% uptime during live classroom pilots
• Enabled data-driven instructor feedback through scalable backend services

Software Engineering Intern – Strategic Operations

• Built a production-grade AI-powered LMS with GPT-based auto-grading
• Reduced manual grading and ops workflows using AWS Lambda, Redis, async job pipelines
• Shipped features with sub-250ms API response times by optimizing PostgreSQL queries and backend logic

Software Engineering Intern – Nowhere Collective

• Launched a digital marketplace serving 100+ users and 50+ vendors
• Built ranking and matching logic using React, TypeScript, SQL
• Improved backend performance, reliability, and CI/CD workflows on AWS

Java • Spring Boot • PostgreSQL • AWS

• Automated quiz creation and grading using GPT
• Designed for correctness, reliability, and real classroom usage

React • Go • SQL • WebSockets

• Real-time instructor dashboards and student interaction tools
• Built scalable analytics pipelines to support teaching at scale

React • TypeScript • SQL • AWS

• Implemented ranking and matching systems
• Focused on clean UX, uptime, and performance under load

Languages: TypeScript, Java, Python, Go, SQL, C++
Frontend: React, Next.js, WebSockets
Backend: Node.js, Express, Spring Boot, REST, gRPC
AI / Data: GPT APIs, agent evaluation pipelines, PostgreSQL, Redis
Infra: AWS (EC2, S3, RDS, Lambda), Docker, CI/CD, GitHub Actions

• I move fast but I don’t ship things I wouldn’t trust myself
• I care about systems that explain themselves, especially when AI is involved
• I am comfortable jumping across frontend, backend, infra, and AI in the same day
• I do my best work in early-stage, high-ownership environments
• I also have a newfound obsession with running, so might grad you to run sooner or later

• Website: https://yashajaysawant.framer.website
• LinkedIn: https://linkedin.com/in/yash-ajay-sawant