I'm an AI researcher focused on large language model inference, transformer memory systems, and efficient long-context evaluation. I build research systems that bridge algorithmic ideas with practical GPU and inference engineering, emphasizing reproducibility and open implementations.
My current research includes DKV (Differential KV Cache Compression), a training-free approach to reducing KV-cache memory through anchor representations, low-rank differential reconstruction, and exact residual preservation. DKV combines compression techniques with optimized inference runtimes and is under review at IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI).
I'm also developing CRBench, a method-agnostic, resource-aware framework for evaluating long-context LLM systems across contextual quality, memory efficiency, and inference performance. The current release is an initial research release, and the full version has been submitted to ASPLOS 2027 in the September review cycle.
I am currently a Research Intern with the Data Systems Group (DSG) at MPI-SWS, hosted by Dr. Laurent Bindschaedler, investigating incremental LLM state maintenance and KV-cache behavior under sequential updates, using controlled experiments to compare cached state against clean reconstruction.
Beyond these projects, I work on LLM inference optimization, GPU systems, transformer memory architectures, model evaluation, and AI benchmarking. I enjoy taking ideas from first principles, validating them experimentally, and turning them into systems that others can reproduce and extend.
Currently exploring incremental LLM state maintenance, adaptive KV-cache compression, memory-efficient transformer inference, long-context evaluation, and GPU-optimized LLM serving.
I believe research is more useful when it can be reproduced, inspected, and extended. My projects aim to provide complete implementations, benchmarks, documentation, and reproducible experiments wherever possible.
- ResearchGate: https://www.researchgate.net/profile/Om-Chimurkar
- Google Scholar: https://scholar.google.com/citations?user=7NXAM-wAAAAJ
- ORCID: https://orcid.org/0009-0004-0518-4598
- LinkedIn: https://www.linkedin.com/in/om-chimurkar
- Email: omchimurkar45@gmail.com