Kumar Shivendu
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKumar Shivendu is a backend and DevOps-focused software engineer with 7 years of experience based in San Francisco, specializing in search and distributed systems. He contributes to production-grade vector search infrastructure at Qdrant, adding robust versioning, health checks, continuous profiling, and shard/segment optimizations for large-scale deployments. Kumar has also implemented secure authentication primitives like TOTP in SuperTokens and enhanced benchmarking tooling for ANN libraries, demonstrating a strong blend of systems design and performance engineering. Comfortable across backend stacks and observability tooling, he leans into practical reliability improvements that make debugging and scaling easier in real-world clusters. An IIT Bhilai alumnus, he pairs academic rigor with hands-on open-source contributions that influence popular projects in the vector search and auth ecosystems.
7 years of coding experience
B.Tech, B.Tech at Indian Institute of Technology, Bhilai
Spanish, Hindi, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (32)
github-ci10
docker10
python10
data-science10
databases10
machine-learning10
authentication10
user-authentication10
vector-database10
dockers10
java10
cicd10
numpy10
javas10
rust10
Programming languages (19)
Github contributions (5)
supertokens/supertokens-core
Jan 2023 - Jan 2023
Open source alternative to Auth0 / Firebase Auth / AWS Cognito
Role in this project:
Back-end Developer
Back-end DeveloperContributions:229 reviews, 1 commit, 17 PRs in 1 day
Contributions summary:Kumar primarily focused on implementing and improving Time-based One-Time Password (TOTP) functionality within the SuperTokens authentication framework. They added in-memory TOTP classes, implemented database queries, and integrated Java-OTP for code generation. The user also improved testing, added cron jobs for expired code removal, and made updates to the TOTP recipe to manage user devices. In addition, the user addressed database constraints and refactored code for efficiency.
email-password-loginsocial-loginkeycloakawsaws-auth
qdrant/qdrant
May 2023 - Apr 2025
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
Role in this project:Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:100 reviews, 84 PRs, 233 pushes in 1 year 10 months
Contributions summary:Kumar contributed to enhancing the Qdrant vector database's functionality, focusing on improvements related to health checks and Docker image builds. They introduced the inclusion of the Git commit ID within the health check endpoint and Docker images, ensuring versioning accuracy. Further contributions involved refactoring and restructuring OpenAPI tests, and integrating Pyroscope for continuous profiling, demonstrating a strong focus on monitoring and debugging. The user also addressed issues related to shard/collection status, and optimizing segments.
pythonapproximate-nearest-neighbor-searchhnswsimilarity-searchvector
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.