Sounak Chakraborty
Senior Software Architect at Activeloop
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolSounak Chakraborty is a Senior Software Architect with nine years of focused experience architecting cloud-native data lakes and large-scale distributed database systems. He has led design and implementation of database internals, query optimizers, and multi-tier cache/workload management for OLAP engines across startups and enterprises, including Oracle, DataPelago, and Activeloop. At DataPelago he co-designed hardware-software optimizations to pursue 100x performance/TCO improvements, and at Activeloop he has contributed backend and MLOps work to the popular Deeplake project, enhancing vector index maintenance for production-grade similarity search. His hands-on expertise spans Spark, Presto, Teradata, Cassandra and CarbonData compaction internals, reflecting deep proficiency in both software and system-level performance tuning. Based in Bengaluru, he combines systems-level thinking with pragmatic engineering to turn complex data infrastructure challenges into scalable, maintainable solutions.
9 years of coding experience
17 years of employment as a software developer
Burnpur Riverside SchoolBachelor of Technology (BTech), Information Technology, Bachelor of Technology (BTech), Information Technology at International Institute of information technology
Github Skills (24)
python10
big-data10
machine-learning10
vector-database10
java10
ml10
javas10
mlops10
index-management10
api-design9
hadoop9
apache9
data-structure9
algorithm9
data-formats9
Programming languages (3)
Github contributions (5)
apache/carbondata
Jan 2017 - Jul 2018
High performance data store solution
Role in this project:
Back-end Developer
Back-end DeveloperContributions:41 commits, 62 PRs, 404 comments in 1 year 6 months
Contributions summary:Sounak focused on enhancing the horizontal compaction of update and delete delta files within the Apache CarbonData project. This involved modifications to Java code, particularly in `CarbonDataMergerUtil.java`, which suggests involvement in the core data merging and compaction processes. The changes included updating table status files, supporting IUD update delta merge status, and implementing functionality related to compaction based on IUD (insert, update, delete) operations.
carbondatadata-formatdata-storeapachebig-data
activeloopai/deeplake
Jul 2023 - Aug 2024
Database for AI. Store Vectors, Images, Texts, Videos, etc. Use with LLMs/LangChain. Store, query, version, & visualize any AI data. Stream data in real-time to PyTorch/TensorFlow. https://activeloop.ai
Role in this project:Back-end Developer & MLOps Engineer
Back-end Developer & MLOps EngineerContributions:30 reviews, 15 PRs, 90 pushes in 1 year 1 month
Contributions summary:Sounak's commits focus on the implementation and maintenance of index management features within the deeplake project. Their work involved modifying the dataset and tensor classes to support the creation, regeneration, and incremental maintenance of vector database indexes, which are essential for efficient similarity searches. They addressed index creation, deletion, and update functionalities, ensuring their correct behavior across different operations, including append, update, and delete, thus contributing to the system's scalability and query performance. They also refactored the code to accommodate changes from external dependencies like Indra and libdeeplake, emphasizing the project's commitment to performance and efficiency.
pythonreal-time-datadata-streamjupyter-notebookvisualize
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.