Want to know if this job is worth applying to?
תל־אביב–יפו, מחוז תל אביב, ישראל
Onsite
Eon is building the next generation of cloud data protection, transforming cloud backups into accessible, searchable, AI-ready data assets. We combine advanced data systems, storage technologies, and AI to help enterprises unlock the value of their backed-up data.
As a Senior Research Scientist, AI & Data Systems at Eon Research, you will conduct applied research at the intersection of AI, data systems, and large-scale storage. Working closely with Assaf Natanzon, VP and Chief Architect, you will explore challenging technical problems, publish research in leading peer-reviewed venues, file patents, and translate research breakthroughs into production-ready product capabilities.
This role is ideal for a researcher with a strong publication record, deep expertise in a relevant research area, and experience bringing research into real-world products.
Query Processing and Query Engines: Query optimization and execution over columnar and lakehouse formats, including Apache Iceberg and Parquet, as well as approximate query processing using sketches and synopses.
Storage Systems: Deduplication, compression, and data layout optimization for large-scale storage.
Data Integration and Discovery: Entity resolution, schema matching, table joinability, and data discovery, including the use of LLMs as judges.
Information Retrieval and RAG: Retrieval-augmented generation over enterprise data, including emails, documents, and databases.
LLM Evaluation and Benchmarking: Evaluation frameworks and benchmarks for LLM agents working with enterprise data.
LLMs for Software Engineering: CI failure diagnosis, automated debugging, and program repair.
PhD in Computer Science or a closely related field. The degree must be completed; candidates currently pursuing a PhD will not be considered.
A deep, sustained research track record in at least one of the relevant areas listed above.
Publications in strong peer-reviewed conferences, such as VLDB, SIGMOD, ICDE, SIGIR, CIKM, FAST, OSDI, USENIX ATC, ACL, EMNLP, NeurIPS, ICML, ICLR, ICSE, or FSE, depending on the research area.
At least 5 years of industry research experience, including work that has been deployed in production.
Hands-on experience with LLMs, including prompting, fine-tuning techniques such as LoRA, model evaluation, and running batch inference on GPUs.
Strong Python skills and the ability to independently design, implement, and execute end-to-end experiments.
Excellent written English and the ability to communicate complex research findings clearly.
Experience in storage, backup, data protection, or enterprise data management.
Research or engineering experience with query engines, lakehouse formats, or approximate query processing.
Experience with cloud platforms such as AWS, Azure, or GCP.
Patents or a track record of translating research into intellectual property.
Experience working with enterprise data sources, including Microsoft 365, email, documents, and databases.
Lead applied research at the intersection of AI, data systems, information retrieval, and large-scale storage.
Identify and investigate challenging technical problems related to making enterprise backup data more searchable, accessible, and useful for AI applications.
Design, implement, and evaluate research ideas through hands-on experimentation, prototyping, and benchmarking.
Develop novel approaches to query processing, data discovery, entity resolution, retrieval, and LLM-powered systems.
Collaborate closely with Assaf Natanzon, VP and Chief Architect, and Eon's engineering and product teams to translate research findings into production-ready capabilities.
Publish research in leading peer-reviewed conferences and contribute to patent applications.
Build evaluation frameworks and benchmarks to measure the quality, accuracy, and performance of AI systems operating on enterprise data.
Stay current with advances in AI and data systems, identify promising research directions, and turn them into practical innovations for Eon's platform