About
I am a Staff Software Engineer on MongoDB Atlas Search, working on vector search and query performance. Previously, I built enterprise Generative AI and machine learning infrastructure at Visa and worked on database query optimization and ML-driven workload management for Amazon Redshift. I earned my Ph.D. in Computer Science from the University of California, San Diego, advised by Prof. Yannis Papakonstantinou.
Vector Search / HNSWAI Retrieval InfrastructureGenerative AI PlatformsQuery ExecutionDistributed SystemsPerformance Engineering
Experience & education
AI Infrastructure · Search & Retrieval · Distributed Systems
Chunbin Lin, Ph.D.
Staff Software Engineer at MongoDB
I build high-performance AI retrieval, vector search, and distributed data infrastructure — bridging deep systems research with production engineering at scale.
Experience at a Glance
MongoDB · Staff Software Engineer
Atlas Search — vector search, indexing, and query performance.
Visa · AI / ML Infrastructure
Enterprise Generative AI platform and machine learning systems.
Amazon Web Services · Redshift
Database query optimization and intelligent workload management.
UC San Diego · Ph.D. in Computer Science
Research in database systems, search, and efficient analytics.
Selected Engineering Projects
A decade of work across AI infrastructure, information retrieval, machine learning platforms, and database systems.
Vector Search & AI Retrieval
Design and optimize vector search infrastructure for large-scale, high-dimensional similarity search and retrieval.
- HNSW index construction, maintenance, merge policies, and memory-aware graph building.
- Query execution and ID lookup optimization for search workloads.
- Nested vector search with hierarchical retrieval, parent filtering, and score aggregation.
HNSW · ANN · Query Optimization · Indexing
Enterprise GenAI Platform
Built infrastructure to serve and govern enterprise applications using multiple large language model providers.
- Integrated OpenAI, Anthropic, and open-source LLMs through a unified platform.
- Developed load balancing, rate limiting, and content inspection services.
- Focused on low-latency, reliable, highly available model access.
LLM Infrastructure · Model Integration · Reliability
Machine Learning Platform
Developed scalable ML infrastructure supporting the model lifecycle from feature engineering to production deployment.
- Feature creation, model training, inference, and deployment.
- AutoML workflows to simplify development and optimization.
- Scalable services for enterprise ML workloads.
ML Platforms · AutoML · MLOps
Database Query Optimization
Improved analytical database performance through query processing and intelligent workload management.
- Join optimization and result cache enhancements.
- Intelligent workload management with ML-based predictions of execution time and resource requirements.
- Resource-aware query scheduling for large analytical workloads.
Database Internals · Query Processing · ML-Based WLM
Engineering Impact
Selected results from controlled performance benchmarks for MongoDB ID lookup optimization. These figures describe specific test configurations, not guaranteed production improvements.
5.5×Latency ImprovementID lookup benchmark
~2.75×Concurrent throughput16-thread benchmark
Performance figures are from internal experiments and should be reviewed for external-publication approval before deploying this page publicly.
Publications
Selected Conference Papers
| C23 |
Auto-WLM: ML-enhanced workload management in Amazon Redshift Gaurav Saxena, Mohammad Rahman, Naresh Chainani, Chunbin Lin, George Caragea, Fahim Chowdhury, Ryan Marcus, Tim Kraska, Ippokratis Pandis, Balakrishnan (Murali) Narayanaswamy Proceedings of ACM Conference on Management of Data (SIGMOD), 2023 Link PDF |
| C22 |
Multivariate Time Series Data Imputation Using Attention-Based Mechanism Jingqi Zhao, Chuitian Rong, Chunbin Lin, Xin Dang Neurocomputing, 2023 |
| C21 |
Highly Efficient String Similarity Search and Join over Compressed Indexes Guorui Xiao, Jin Wang, Chunbin Lin, Carlo Zaniolo IEEE International Conference on Data Engineering (ICDE), 2022 |
| C20 |
Workload-Aware Performance Tuning for Autonomous DBMSs Zhengtong Yan, Jiaheng Lu, Naresh Chainani, Chunbin Lin IEEE International Conference on Data Engineering (ICDE), 2021 Link |
| C19 |
Evaluating List Intersection on SSDs for Parallel I/O Skipping Jianguo Wang, Chunbin Lin, Yannis Papakonstantinou, Steven Swanson IEEE International Conference on Data Engineering (ICDE), 2021 |
| C18 |
Plato: Approximate Analytics over Compressed Time Series with Tight Deterministic Error Guarantees Chunbin Lin, Etienne Boursier, Yannis Papakonstantinou Proceedings of the VLDB Endowment (PVLDB), 2020 Link PDF |
| C17 |
Fast Error-tolerant Location-aware Query Autocompletion Jin Wang, Chunbin Lin IEEE International Conference on Data Engineering (ICDE), 2020 Link PDF |
| C16 |
Motif Discovery Using Similarity-Constraints Deep Neural Networks Chuitian Rong, Ziliang Chen, Chunbin Lin, Jianming Wang International Conference on Database Systems for Advanced Applications (DASFAA), 2020 Link PDF |
| C15 |
Synergy of Database Techniques and Machine Learning Models for String Similarity Search and Join Jiaheng Lu, Chunbin Lin, Jin Wang, Chen Li Proceedings of the 28th ACM International Conference on Information and Knowledge Management (CIKM), pages: 2975-2976, 2019 Link PDF Full Version (PDF) Website |
| C14 |
MF-Join: Efficient Fuzzy String Similarity Join with Multi-level Filtering Jin Wang, Chunbin Lin, Carlo Zaniolo IEEE International Conference on Data Engineering (ICDE), 2019 Link PDF |
| C13 |
Scalable Metric Similarity Join Using MapReduce Jiacheng Wu, Yong Zhang, Jin Wang, Chunbin Lin, Yingjia Fu, Chunxiao Xing IEEE International Conference on Data Engineering (ICDE), 2019 Link |
| C12 |
An Efficient Sliding Window Approach for Approximate Entity Extraction with Synonyms Jin Wang, Chunbin Lin, Mingda Li, Carlo Zaniolo International Conference on Extending Database Technology (EDBT), 2019 Link PDF |
| C11 |
An Experimental Study of Bitmap Compression vs. Inverted List Compression Jianguo Wang, Chunbin Lin, Yannis Papakonstantinou, Steven Swanson Proceedings of ACM Conference on Management of Data (SIGMOD), pages: 993-1008, 2017 Link PDF |
| C10 |
MILC: Inverted List Compression in Memory Jianguo Wang, Chunbin Lin, Ruining He, Moojin Chae, Yannis Papakonstantinou, Steven Swanson Proceedings of the VLDB Endowment (PVLDB), Volume 10, Issue 8, pages: 853-864, 2017 Link PDF |
| C9 |
Towards heterogeneous keyword search Chunbin Lin, Jianguo Wang, Chuitian Rong Proceedings of the ACM Turing 50th Celebration Conference-China (ACM TUR-C), pages: 1-6, 2017 Link PDF |
| C8 |
Fast and Scalable Distributed Set Similarity Joins for Big Data Analytics Chuitian Rong, Chunbin Lin, Yasin Silva, Jianguo Wang, Wei Lu, Xiaoyong Du Proceedings of the International Conference on Data Engineering (ICDE), pages: 1059-1070, 2017 Link PDF |
| C7 |
Answer yes/no queries in search engines Chunbin Lin The Conference on Innovative Data Systems Research (CIDR), 2017 |
| C6 |
Fast In-Memory SQL Analytics on Typed Graphs Chunbin Lin, Benjamin Mandel, Yannis Papakonstantinou, Matthias Springer Proceedings of the VLDB Endowment (PVLDB), Volume 10, Issue 3, pages: 265-276, 2016 Link PDF Website |
| C5 |
HippogriffDB: Balancing I/O and GPU Bandwidth in Big Data Analytics Jing Li, Hung-Wei Tseng, Chunbin Lin, Yannis Papakonstantinou, Steven Swanson Proceedings of the VLDB Endowment (PVLDB), Volume 9, Issue 14, pages: 1647-1658, 2016 Link PDF |
| C4 |
Sherlock: Sparse Hierarchical Embeddings for Visually-aware One-class Collaborative Filtering Ruining He, Chunbin Lin, Jianguo Wang, Julian McAuley Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), pages: 3740-3746, 2016 Link PDF |
| C3 |
String similarity measures and joins with synonyms Jiaheng Lu, Chunbin Lin, Wei Wang, Chen Li, Haiyong Wang Proceedings of ACM Conference on Management of Data (SIGMOD), pages: 373-384, 2013 Link PDF |
| C2 |
Processing XML Twig Pattern Query with Wildcards Huayu Wu, Chunbin Lin, Tok Wang Ling, Jiaheng Lu International Conference on Database and Expert Systems Applications (DEXA), pages: 326-341, 2012 Link |
| C1 |
Optimal top-k generation of attribute combinations based on ranked lists Jiaheng Lu, Pierre Senellart, Chunbin Lin, Xiaoyong Du, Shan Wang, Xinxing Chen Proceedings of ACM Conference on Management of Data (SIGMOD), pages: 409-420, 2012 Link PDF |
Selected Journal Articles
(* denotes corresponding author)| J3 |
Boosting Approximate Dictionary-based Entity Extraction with Synonyms Jin Wang, Chunbin Lin*, Mingda Li, Carlo Zaniolo Information Sciences, Volume 530, pages: 1-21, 2020. (Impact Factor: 5.524) Link |
| J2 |
Optimal Algorithms for Selecting Top-k Combinations of Attributes: Theory and Applications Chunbin Lin*, Jiaheng Lu, Zhewei Wei, Jianguo Wang, Xiaokui Xiao The International Journal on Very Large Data Bases (VLDB Journal), Volume 27, Issue 1, pages: 27-52, 2018 Link PDF |
| J1 |
Boosting the Quality of Approximate String Matching by Synonyms Jiaheng Lu, Chunbin Lin, Wei Wang, Chen Li, Xiaokui Xiao ACM Transactions on Database Systems (TODS), Volume 40, Issue 3, pages: 1-42, 2015 Link PDF |
Selected Demo papers
| D5 |
GQFast: Fast Graph Exploration with Context-Aware Autocompletion Chunbin Lin, Jianguo Wang, Yannis Papakonstantinou Proceedings of the International Conference on Data Engineering (ICDE), pages 1389-1390, 2017 Link Demo Video |
| D4 |
SpiderX: Fast XML Exploration System Chunbin Lin, Jianguo Wang Proceedings of International World Wide Web Conference (WWW), pages: 237-241, 2017 Link PDF |
| D3 |
Location-sensitive Query Auto-completion Chunbin Lin, Jianguo Wang, Jiaheng Lu Proceedings of International World Wide Web Conference (WWW), pages: 819-820, 2017 Link PDF |
| D2 |
Fashionista: A Fashion-aware Graphical System for Exploring Visually Similar Items Ruining He, Chunbin Lin, Julian McAuley Proceedings of the International Conference on World Wide Web (WWW), pages 199-202, 2016 Link PDF |
| D1 |
LotusX: A Position-Aware XML Graphical Search System with Auto-Completion Chunbin Lin, Jiaheng Lu, Tok Wang Ling, Bogdan Cautis Proceedings of the International Conference on Data Engineering (ICDE), pages 1265-1268, 2012 Link PDF |
Academic Service
|
|
|