|
Research
I build data systems for the AI era, with research spanning two complementary paths:
Vector search.
I design and evaluate vector search techniques inside modern database systems, with an emphasis on
filtered and filter-agnostic vector search, high-dimensional embedding operators, and indexes that
integrate efficiently with relational query processing and stream processing.
Declarative agentic AI data systems.
I develop declarative abstractions and data-system primitives for building, orchestrating, and
optimizing agent harnesses and LLM workflows, enabling systems to plan and adapt multi-step AI
workloads.
|
|
Publications
|
Credo: Reusable Declarative Primitives for Agentic Workloads
Duo Lu, Andrew Crotty, Uğur Çetintemel
The Conference on Innovative Data Systems Research (CIDR), 2027, To be appear.
|
Declarative Data Services: Structured Agentic Discovery for Composing Data Systems
Shanshan Ye*, Duo Lu*
AI Agents for Discovery in the Wild (AID-WilD)
@ ACM Conference on AI and Agentic System (ACM CAIS), 2026.
* Equal contribution.
Poster
|
Credo: Declarative Control of LLM Pipelines via Beliefs and Policies
Duo Lu, Andrew Crotty, Uğur Çetintemel
International Conference on Very Large Data Bases (VLDB), 2026. (Demo Track)
Slides /
Poster /
Code
|
An In-Depth Study of Filter-Agnostic Vector Search on a PostgreSQL Database System
Duo Lu, Helena Caminal, Manos Chatzakis, Yannis Papakonstantinou, Yannis Chronis, Vaibhav Jain, Fatma Ozcan
International Conference on Management of Data (SIGMOD), 2026.
Presentation /
Poster /
Code /
BDA 2026
|
Making Prompts First-Class Citizens for Adaptive LLM Pipelines
Uğur Çetintemel, Shu Chen, Alexander W. Lee, Deepti Raghavan, Duo Lu, Andrew Crotty
The Conference on Innovative Data Systems Research (CIDR), 2026.
|
VectraFlow: Integrating Vectors into Stream Processing
Duo Lu, Siming Feng, Jonathan Zhou, Franco Solleza, Malte Schwarzkopf, Uğur Çetintemel
The Conference on Innovative Data Systems Research (CIDR), 2025.
Featured in ANNS workshop at EDBT 2026
|
Evaluating Persistent Memory Range Indexes: Part Two
Yuliang He, Duo Lu, Kaisong Huang, Tianzheng Wang
International Conference on Very Large Data Bases (VLDB), 2022.
Extended Version /
Slides /
Poster /
Code
|
|
Experience
|
Jun 2026 ~ Aug 2026
Student Researcher
Sunnyvale, CA, United States
 Google, Systems Research Group ( SRG)
Hosts: Helena Caminal, Yannis Chronis, Fatma Özcan
Developing a hybrid search system that fuses retrievers for natural-language queries over relational tables.
|
May 2025 ~ Dec 2025
Student Researcher
Sunnyvale, CA, United States
 Google, Systems Research Group ( SRG)
Hosts: Helena Caminal, Yannis Chronis, Fatma Özcan
Developed and evaluated filtered vector search indexes in collaboration with the Google AlloyDB team.
|
May 2022 ~ Aug 2023
Embedded Software Developer
Vancouver, BC, Canada
 Fortinet, FortiOS Routing Team
Manager: Henry Xing
Maintained and improved routing protocols such as OSPF and BGP in FortiOS.
|
|
Talks
|
Sep 2026
Intelligent Search: Making Multi-Column Retrieval Practical
Remote
Invited Talk, Google, Vector Search TL Council
Host: Prof. Jianguo Wang
|
Apr 2026
Filtered Vector Search in PostgreSQL
Host: Prof. Jianguo Wang
|
|
Services
|
Shadow Program Committee, VLDB 2027
Reviewer: NOVAS 2026 @ VLDB 2026, AID-WilD 2026 @ ACM CAIS 2026
Artifact Evaluation Committee, ACM CAIS 2026
|
Power by Github Pages - Template from jonbarron
|
|