Sanket Kedia
Founding Distributed Systems Engineer at Chroma
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSanket Kedia is a founding distributed systems engineer based in San Francisco with about a decade of experience building low-latency, real-time systems and core database infrastructure. He holds a dual B.Tech+M.Tech in Computer Science from IIT Kharagpur and has shipped production services at companies ranging from Oracle and Yugabyte to Netflix and the AI-native embedding DB Chroma. Sanket’s expertise spans Rust, modern C++, and Java Spring Boot, with hands-on work on HNSW indexes, full-text and metadata engines, and single-digit-millisecond KV systems. He contributed OpenTelemetry tracing and end-to-end span propagation to Chroma’s query and compaction services, enabling observability across complex query orchestration and worker threads. A researcher at heart, his early academic work includes GPU kernel optimization and published NLP research, reflecting a pattern of turning deep systems knowledge into practical, measurable improvements. Sanket combines low-level systems craftsmanship with production reliability and a proven ability to introduce cross-team observability and performance enhancements.
10 years of coding experience
7 years of employment as a software developer
Matriculation Science, Matriculation Science at M.C.Kejriwal VidyapeethHigher Secondary Science, Higher Secondary Science at Adwaita Mission High SchoolDual (B.Tech + M.Tech) Computer Science and Engineering, Dual (B.Tech + M.Tech) Computer Science and Engineering at Indian Institute of Technology, Kharagpur
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (21)
c-language10
tracer10
back-end-development10
databases10
load-balancing10
distributed-systems10
database-design10
opentelemetry10
rust10
cprogramming-language10
trace10
distributed-database10
devops10
relational-databases10
sql-database10
Programming languages (3)
Github contributions (5)
chroma-core/chroma
Apr 2024 - Mar 2025
the AI-native open-source embedding database
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:401 reviews, 253 PRs, 730 pushes in 11 months
Contributions summary:Sanket implemented end-to-end tracing for the Query Service, specifically for the `query_vectors()` RPC. They integrated span inheritance and propagation across components, including the HNSW Query orchestrator and worker threads. Furthermore, the user added OpenTelemetry (OTEL) support to the query and compaction services, enabling the export of traces and spans in an OpenTelemetry format compatible with backends such as Jaeger. This included modifying the config for the service, adding a new tracing configuration file, and modifying the lib.rs and query_service.rs to enable tracing.
document-retrievalembeddingsllms
sanketkedia/CodeDeployGitDemo
Apr 2019 - Apr 2019
Contributions:9 pushes, 1 branch in 1 day
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.