About
I build the layer nobody notices until it breaks. My job is to make sure it doesn't.
I'm Rabindra Yadav, a Software Development Engineer II at AWS in Austin, Texas. My work sits underneath AWS managed databases: the DNS architecture, IPv6 networking, and availability machinery that roughly 700,000 public databases depend on to answer queries. When that layer is healthy, nobody thinks about it. That's the point.
I grew up in Gaighat, a small town in southeastern Nepal, and came to Texas for college at UT Arlington. Infrastructure at this scale is unforgiving in a way I like. A configuration drift in one region, a throttling limit hit during a traffic spike, a DNS record that resolves differently inside the VPC than outside. Small things become fleet-wide things fast. The work rewards precision, and it punishes hand-waving.
Lately I've been pointing AI at operations. I architected a RAG-based diagnostic tool on Amazon Bedrock that indexes 10,000+ internal technical documents for semantic search, cutting incident investigation time by 74%. On-call is where engineers burn out; making it faster and saner is some of the highest-leverage work I know.
Where I've been
Graduated Dec 2022
B.S. Computer Science, UT Arlington
Arlington, Texas · GPA 3.94 / 4.00
Maverick Academic Scholarship covered 80% of tuition. Won the Cyneta Outstanding Undergraduate Teaching Assistant Award, partly by building a Python grading engine that cut manual grading overhead 40% for 450+ students. First lesson in automation: the best tool is the one that deletes a chore.
May - Aug 2022
SDE Intern, AWS
Lambda Networking · Seattle, Washington
Built a network health insights dashboard that automated error detection and cut ticket resolution time in half. Twelve weeks, one shipped tool, and a return offer.
Jan 2023 - Mar 2025
SDE I → II, AWS
Serverless Compute Networking · Seattle, Washington · promoted Oct 2024
Led a microservice decoupling initiative, moving networking logic into a dedicated gRPC service, saving $3M a year in infrastructure with cold boots 180 seconds faster. Pushed DNS availability from 94% to 97% for Lambda non-VPC environments by root-causing resource exhaustion. Unblocked the GA launch of Xanadu by migrating its legacy networking stack to the secure, SOC2-compliant mainline.
Mar 2025 - present
SDE II, AWS
RDS Platform Networking · Austin, Texas
Orchestrated the dual-stack IPv6 launch across 30+ regions for ~700K public databases. Resolved a fleet-wide DNS API throttling crisis and restored 99.99% availability. Built the AI on-call diagnostic tool. Currently leading faster DNS propagation for Multi-AZ failover, projected to cut propagation time by 70% across millions of customer databases.
Contact
The short version: I make large systems boring, in the best possible way. If you're building something interesting at scale, write to rabindrayadav9596@gmail.com.