Senior Java Acceleration Engineer, Vector Search and Database Acceleration

NVIDIA•Us, NY

About The Position

NVIDIA is hiring engineers to work on RAPIDS, a suite of open source software libraries that accelerate end-to-end data processing and vector search algorithms on GPUs. RAPIDS relies on NVIDIA CUDA for low-level compute optimization, but exposes that high-performance GPU compute through user-friendly languages like Java. We’re expanding our vector search and database acceleration to include experienced Java engineers. This team builds next-generation building blocks for accelerating Java-based libraries like Lucene and JVector, which are used in widely popular databases like OpenSearch, Solr, and Elasticsearch. You will also work to accelerate popular analytics platforms like Presto. In this role, you will develop, benchmark, and explore novel tuned custom solutions for accelerating vector preprocessing, indexing, and search. This includes end-to-end database acceleration and scale, including introducing scalable architectural improvements, optimizing disk-based indexing and using next-generation hardware to benchmark data volumes not tractable with today’s CPU-centric solutions. This is a great chance to take advantage of your Java CUDA/C++ skills and make a huge impact across a rapidly growing industry. Vector search is an exciting new field at the intersection of data processing and AI. You’ll work closely with the RAPIDS team of stellar engineering redefining what’s possible in the world of unstructured information retrieval.

Requirements

BS or MS degree in Computer Science, Computer Engineering, or a related subject area, or equivalent experience.
12+ years of validated experience programming in Java within NoSQL DBs: Lucene, Elasticsearch, OpenSearch, MongoDB, Solr
Experience with Vector Databases: Milvus, Pinecone, LanceDB
Strong analytical problem-solving skills, algorithms, and mathematics fundamentals.
Excellent software development skills: programming, debugging, performance analysis, and test design, especially within the Java ecosystem and the JVM
Ability to work independently and manage your own development efforts.
Good communication and documentation habits.

Nice To Haves

Recent experience in developing and implementing distributed algorithms across various platforms, including HPC and cloud environments.
Strong background in building and maintaining distributed systems.
Skilled in debugging complex systems that involve multiple programming languages and hardware configurations.
Knowledge of Nearest Neighbor Algorithms, including graph-based methods and inverted file indexing.
Familiarity with machine learning techniques such as clustering and dimensionality reduction; GPU programming experience is a bonus, with training available for those lacking it.

Responsibilities

Analyze, design, and implement optimized GPU algorithms for large-scale vector search, databases, and machine learning.
Expand and improve the integration of NVIDIA cuVS into relevant high-level vector search libraries and vector databases.
Drive performance analysis, benchmarking, and troubleshooting of associated libraries.
Collaborate with a multi-functional team to understand requirements and implement or improve solutions