Rahul Krishna

Sr. Research Scientist, IBM Research, NY.

imralk AT gmail DOT com

I’m a Senior Research Scientist at IBM Research, where my work sits at the intersection of large language models and program analysis for software engineering. My current research builds agentic systems that bridge code and IT operations — spanning root-cause analysis of cloud incidents (Praxis), automated test generation (ASTER and SAINT), code translation and modernization (SpecTra, Lost in Translation, and C2SaferRust), and the open-source CodeLLM-DevKit (CLDK) framework, a cross-language middleware layer that grounds agents with program-analysis context.

Earlier work spans AI-guided microservice migration (CARGO), crowdtesting management (iSENSE), and deep-learning vulnerability detection. My research appears regularly at the top venues in software engineering and systems (ICSE, FSE, ASE, OOPSLA, EuroSys, TSE, DSN).

My broader interests include software security (particularly vulnerability detection and fuzzing) and causal inference (applied to diagnosing and repairing performance faults and reasoning about configurable-system behavior) — themes from my postdoc at Columbia University. Earlier still, I completed a Ph.D. at NC State University on actionable analytics and transfer learning for software engineering.

Resume.
Publications.

news

Jun 1, 2026	🏆 Praxis won the Best Paper Award at DSN’26 — agentic root cause analysis for code-related incidents in cloud applications.
May 7, 2026	New preprint: ScarfBench — a benchmark for cross-framework application migration in enterprise Java. Project page: scarfbench.info.
Mar 1, 2026	Process-Centric Analysis of Agentic Software Systems accepted at OOPSLA’26.
Feb 1, 2026	SAINT accepted at ICSE’26 — service-level integration test generation with program analysis and LLM-based agents.
Jan 1, 2026	C2SaferRust published in IEEE TSE — transforming C projects into safer Rust with neurosymbolic techniques.

Selected Publications

DSN

Praxis: Agentic Structured Graph Traversal for Root Cause Analysis of Code-related Incidents in Cloud Applications

Cui, Shengkun, Krishna, Rahul, Jha, Saurabh, and Iyer, Ravishankar K.

In Proceedings of the 56th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN) 2026

🏆 Best Paper Award

arXiv
ICSE-SEIP

ASTER: Natural and Multi-Language Unit Test Generation with LLMs

Pan, Rangeet, Kim, Myeongsoo, Krishna, Rahul, Pavuluri, Raju, and Sinha, Saurabh

In Proceedings of the 47th IEEE/ACM International Conference on Software Engineering: Software Engineering in Practice (ICSE-SEIP) 2025

🏆 Best Paper Award

HTML PDF
TSE

Deep learning based vulnerability detection: Are we there yet?

Chakraborty, Saikat, Krishna, Rahul, Ding, Yangruibo, and Ray, Baishakhi

IEEE Transactions on Software Engineering. 2021

🏆 Best Paper Award Runner Up

Abs HTML PDF

Automated detection of software vulnerabilities is a fundamental problem in software security. Existing program analysis techniques either suffer from high false positives or false negatives. Recent progress in Deep Learning (DL) has resulted in a surge of interest in applying DL for automated vulnerability detection. Several recent studies have demonstrated promising results achieving an accuracy of up to 95% at detecting vulnerabilities. In this paper, we ask, "how well do the state-of-the-art DL-based techniques perform in a real-world vulnerability prediction scenario?". To our surprise, we find that their performance drops by more than 50%. A systematic investigation of what causes such precipitous performance drop reveals that existing DL-based vulnerability prediction approaches suffer from challenges with the training data (e.g., data duplication, unrealistic distribution of vulnerable classes, etc.) and with the model choices (e.g., simple token-based models). As a result, these approaches often do not learn features related to the actual cause of the vulnerabilities. Instead, they learn unrelated artifacts from the dataset (e.g., specific variable/function names, etc.). Leveraging these empirical findings, we demonstrate how a more principled approach to data collection and model design, based on realistic settings of vulnerability prediction, can lead to better solutions. The resulting tools perform significantly better than the studied baseline: up to 33.57% boost in precision and 128.38% boost in recall compared to the best performing model in the literature. Overall, this paper elucidates existing DL-based vulnerability prediction systems’ potential issues and draws a roadmap for future DL-based vulnerability prediction research. In that spirit, we make available all the artifacts supporting our results: https://github.com/VulDetProject/ReVeal
ICSE

SAINT: Service-level Integration Test Generation with Program Analysis and LLM-based Agents

Pan, Rangeet, Pavuluri, Raju, Huang, Ruikai, Krishna, Rahul, Stennett, Tyler, Orso, Alessandro, and Sinha, Saurabh

In Proceedings of the 48th IEEE/ACM International Conference on Software Engineering (ICSE) 2026

arXiv
ICSE

Lost in Translation: A Study of Bugs Introduced by Large Language Models while Translating Code

Pan, Rangeet, Ibrahimzada, Ali Reza, Krishna, Rahul, Sankar, Divya, Wassi, Lambert Pouguem, Merler, Michele, Sobolev, Boris, Pavuluri, Raju, Sinha, Saurabh, and Jabbarvand, Reyhaneh

In Proceedings of the 46th IEEE/ACM International Conference on Software Engineering (ICSE) 2024

HTML PDF
ASE

CARGO: AI-Guided Dependency Analysis for Migrating Monolithic Applications to Microservices Architecture

Nitin, Vikram, Asthana, Shubhi, Ray, Baishakhi, and Krishna, Rahul

In Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering (ASE ’22), October 10–14, 2022, Ann Arbor, MI, USA. 2022

🏆 Best Paper Award

Abs HTML PDF

Microservices Architecture (MSA) has become a de-facto standard for designing cloud-native enterprise applications due to its efficient infrastructure setup, service availability, elastic scalability, dependability, and better security. Existing (monolithic) systems must be decomposed into microservices to harness these characteristics. Since manual decomposition of large scale applications can be laborious and error-prone, AI-based systems to detect decomposition strategies are gaining popularity. However, the usefulness of these approaches is limited by the expressiveness of the program representation and their inability to model the application’s dependency on critical external resources such as databases. Consequently, partitioning recommendations offered by current tools result in architectures that result in (a) distributed monoliths, and/or (b) force the use of (often criticized) distributed transactions. This work attempts to overcome these challenges by introducing CARGO(short for [C]ontext-sensitive l[A]bel p[R]opa[G]ati[O]n)-a novel un-/semi-supervised partition refinement technique that uses a context- and flow-sensitive system dependency graph of the monolithic application to refine and thereby enrich the partitioning quality of the current state-of-the-art algorithms. CARGO was used to augment four state-of-the-art microservice partitioning techniques that were applied on five Java EE applications (including one industrial scale proprietary project). Experiments demostrate that CARGO can improve the partition quality of all modern microservice partitioning techniques. Further, CARGO substantially reduces distributed transactions and a real-world performance evaluation of a benchmark application (deployed under varying loads) shows that CARGO also lowers the overall the latency of the deployed microservice application by 11% and increases throughput by 120% on average.
ICSE

iSENSE: Completion-aware crowdtesting management

Wang, Junjie, Yang, Ye, Krishna, Rahul, Menzies, Tim, and Wang, Qing

In 2019 IEEE/ACM 41st International Conference on Software Engineering (ICSE) 2019

🏆 Distinguished Paper Award

HTML PDF