Hi, I'm Shiv. This is my sincere-posting blog on computers, art, math and consciousness.
Professional
- Current: SDE 2 @ AWS
- Scaling EKS to 100k+ nodes
- Optimising scheduling and execution of large-scale distributed training jobs
- Solving obscure GPU, driver, kernel etc bugs for large-scale distributed training
- Previously: Lead Machine Learning Engineer @ Glance.
- I built Alchemist, an experimentation platform that:
- operates at ~1ms p99 latency at ~100K QPS
- can alter user experience of 200 million+ users, within minutes
- powers experimentation for over 200 million+ users
I'm great at building high-performance + low-latency ML systems. I write good Python, Go, Triton, and CUDA code.
Education
- MS CS, Georgia Tech, Atlanta - GPA: 3.9/4.0
- BTech CS, VIT, Pune - CGPA: 8.39/10.0
Contact Me
Feel free to reach out to me to talk about literally anything. I love Sparkly People.