Rohit Prajapati

PhD Candidate, IIT Jodhpur  ·  Researcher, Shell R&D

I research streaming graph algorithms and distributed systems, with a focus on how graphs can serve as a unifying framework for real-time analytics, scalable computation, and efficient AI. My work spans algorithm design for dynamic graph problems, distributed frameworks for large-scale graph processing, and the application of graph-structured reasoning to AI systems. At Shell R&D, I build and optimize large-scale LLM inference and retrieval pipelines on HPC infrastructure, working at the boundary between systems research and applied AI.

Rohit Prajapati
Lately
Jul 2026
Presented SAGA at HPDC 2026 in Cleveland, OH.
Jun 2026
Attending SIGMOD/PODS 2026 in Bengaluru.
May 2026
Paper accepted at HPDC 2026 — SAGA: A Framework for State-Aware Streaming Graph Analytics.
Apr 2026
Paper accepted at SIGMOD 2026 — PJsim: Towards Precise and Scalable Graph Similarity.
Jan 2026
Paper accepted at ICDCN 2026 — A Precise and Closed-Form Solution for Edge-Ranking.
Aug 2025
Joined Shell R&D as Researcher in the Digital & Scientific HPC team.
Jan 2025
Started PhD at IIT Jodhpur, SPADE Research Lab.
Publications
HPDC 2026
SAGA: A Framework for State-Aware Streaming Graph Analytics
Rohit Prajapati, Prajjwal Nijhara, Dip Sankar Banerjee  ·  Cleveland, OH, USA
ACM International Symposium on High-Performance Parallel and Distributed Computing
DOI
@inproceedings{prajapati2026saga, author = {Prajapati, Rohit and Banerjee, Dip Sankar}, title = {SAGA: A Framework for State-Aware Streaming Graph Analytics}, booktitle = {Proceedings of the 35th International Symposium on High-Performance Parallel and Distributed Computing}, series = {HPDC '26}, year = {2026}, address = {Cleveland, OH, USA}, publisher = {ACM}, }
SIGMOD 2026
PJsim: Towards Precise and Scalable Graph Similarity
Prajjwal Nijhara, Jainan Tandel, Rohit Prajapati, Dip Sankar Banerjee  ·  Bangalore, India
ACM International Conference on Management of Data
DOI
@inproceedings{nijhara2026pjsim, author = {Nijhara, Prajjwal and Tandel, Jainan and Prajapati, Rohit and Banerjee, Dip Sankar}, title = {PJsim: Towards Precise and Scalable Graph Similarity}, booktitle = {Proceedings of the ACM SIGMOD International Conference on Management of Data}, series = {SIGMOD '26}, year = {2026}, address = {Bangalore, India}, publisher = {ACM}, }
ICDCN 2026
A Precise and Closed-Form Solution for Edge-Ranking
Prajjwal Nijhara, Jainan Tandel, Rohit Prajapati, Dip Sankar Banerjee  ·  Nara, Japan
International Conference on Distributed Computing and Networking
DOI
@inproceedings{nijhara2026edgeranking, author = {Nijhara, Prajjwal and Tandel, Jainan and Prajapati, Rohit and Banerjee, Dip Sankar}, title = {A Precise and Closed-Form Solution for Edge-Ranking}, booktitle = {Proceedings of the International Conference on Distributed Computing and Networking}, series = {ICDCN '26}, year = {2026}, address = {Nara, Japan}, publisher = {ACM}, }