LLM-serving & retrieval platform (ILUP)
The shared serving and retrieval layer product teams build on instead of rolling their own. It backs AI and search across six Adobe products; I own its serving and reliability layers.
Read more →Varun Kotte, Staff ML Engineer, Adobe
I build retrieval and LLM systems that stay reliable in production.
I own the serving and reliability layers of the internal LLM platform behind AI and search in six Adobe products, and I publish research on when a model's output can be trusted.
6Adobe products on the platform
8sole-authored papers
~7×Lightroom search click-through
Highlights
- Seven product teams build on the internal LLM-serving and retrieval platform whose serving and reliability layers I own.
- Replaced five per-team LLM integrations with one shared LLM-as-a-Service pattern, new-model integration dropped from weeks to days.
- Built AutoPrompt, an internal tool that cut prompt-engineering setup from days to hours for engineers across the platform.
- Shipped Lightroom semantic search to general availability; the feature's search click-through rose roughly seven-fold.
- Peer reviewer for ICML, NeurIPS, ACL, and ICLR.
- Named inventor on a published Adobe patent application for closed-domain question answering.
- Sole-authored papers at ANNPR 2026 (oral), CVPR, ICLR, and ICML workshops.
Selected work
AutoPrompt
An internal tool that turns a team's own evaluation set into a working prompt baseline, cutting setup from days to hours.
Read more →Lightroom semantic search
The named-entity layer behind intent-based photo search; click-through rose roughly seven-fold at launch.
Read more →