World models enable agents to anticipate the effects of their actions by internalizing environment dynamics. In enterprise systems, however, these dynamics are often defined by tenant-specific business logic that varies across deployments and evolves over time, making models trained on historical transitions brittle under deployment shift. We ask a question the world-models literature has not addressed: when the rules can be read at inference time, does an agent still need to learn them? We argue, and demonstrate empirically, that in settings where transition dynamics are configurable and readable, runtime discovery complements offline training by grounding predictions in the active system instance. We propose enterprise discovery agents, which recover relevant transition dynamics at runtime by reading the system’s configuration rather than relying solely on internalized representations. We introduce CascadeBench, a reasoning-focused benchmark for enterprise cascade prediction that adopts the evaluation methodology of World of Workflows on diverse synthetic environments, and use it together with deployment-shift evaluation to show that offline-trained world models can perform well in-distribution but degrade as dynamics change, whereas discovery-based agents are more robust under shift by grounding their predictions in the current instance. Our findings suggest that, in configurable enterprise environments, agents should not rely solely on fixed internalized dynamics, but should incorporate mechanisms for discovering relevant transition logic at runtime.
@article{nair2026enterprisesystemsneedlearned,title={Do Enterprise Systems Need Learned World Models? The Importance of Context to Infer Dynamics},author={Sethumadhavan Nair, Jishnu and Bechard, Patrice and Maheshwary, Rishabh and Dasgupta, Surajit and Ramachandran, Sravan and Bhagat, Aakash and Radhakrishna, Shruthan and Pattnaik, Pulkit and Obando-Ceron, Johan and Malay, Shiva Krishna Reddy and Davasam, Sagar and Subramanian, Seganrasan and Mittal, Vipul and Nemala, Sridhar Krishna and Pal, Christopher and Sunkara, Srinivas and Rajeswar, Sai},year={2026},month=jul,eprint={2605.12178},archiveprefix={arXiv},primaryclass={cs.AI},journal={NeurIPS 2026: The Fortieth Annual Conference on Neural Information Processing Systems (Main Track)},}
ICML
EnterpriseOps-Gym: Environments and Evaluations for Stateful Agentic Planning and Tool Use in Enterprise Settings
Shiva Krishna Reddy Malay, Shravan Nayak, Jishnu Sethumadhavan Nair, Sagar Davasam, and 5 more authors
International Conference on Machine Learning (ICML), Mar 2026
EnterpriseOps-Gym is a large-scale benchmark for evaluating the agentic planning and tool-use capabilities of LLM agents across enterprise operations. It comprises 1,150 expert-curated tasks spanning 8 enterprise domains, each running against live containerized MCP servers backed by realistic, fully synthetic databases. Unlike static QA benchmarks, EnterpriseOps-Gym evaluates agents on final environment state using SQL verifiers - meaning agents are rewarded for achieving the correct outcome, not for following a rigid action sequence. Tasks require long-horizon multi-step reasoning, strict policy compliance, and precise tool invocation under complex data dependencies.
@article{malay2026enterpriseopsgymenvironmentsevaluationsstateful,title={EnterpriseOps-Gym: Environments and Evaluations for Stateful Agentic Planning and Tool Use in Enterprise Settings},author={Malay, Shiva Krishna Reddy and Nayak, Shravan and Sethumadhavan Nair, Jishnu and Davasam, Sagar and Tiwari, Aman and Madhusudhan, Sathwik Tejaswi and Nemala, Sridhar Krishna and Sunkara, Srinivas and Rajeswar, Sai},journal={International Conference on Machine Learning (ICML)},year={2026},month=mar,eprint={2603.13594},archiveprefix={arXiv},primaryclass={cs.AI},}
Patent
Smart Persistence of Model for Effective Predictions and Updates
Sriram Puttagunta, Jishnu Sethumadhavan Nair, Bidyapati Pradhan, Nirali Dineshbhai Popat, and 4 more authors
Jan 2026
US Patent 20250315447 B1 (Issued: January 27, 2026)
@misc{patent_smart_persistence2026,title={Smart Persistence of Model for Effective Predictions and Updates},author={Puttagunta, Sriram and Sethumadhavan Nair, Jishnu and Pradhan, Bidyapati and Popat, Nirali Dineshbhai and Ramachandran, Sravan and Mittal, Vipul and Subramanian, Seganrasan and Chenna, Ranga Prasad},year={2026},month=jan,day={27},note={US Patent 20250315447 B1 (Issued: January 27, 2026)},publisher={ServiceNow Inc},}
2025
AAAI
DNR Bench: Benchmarking Over-Reasoning in Reasoning LLMs
Masoud Hashemi, Oluwanifemi Bamgbose, Sathwik Tejaswi Madhusudhan, Jishnu Sethumadhavan Nair, and 2 more authors
AAAI Conference on Artificial Intelligence 2026, Jan 2025
Benchmarking over-reasoning capabilities in large language models designed for reasoning tasks.
@article{dnrbench2025,title={DNR Bench: Benchmarking Over-Reasoning in Reasoning LLMs},author={Hashemi, Masoud and Bamgbose, Oluwanifemi and Madhusudhan, Sathwik Tejaswi and Sethumadhavan Nair, Jishnu and Tiwari, Aman and Yadav, Vikas},journal={AAAI Conference on Artificial Intelligence 2026},year={2025},}
arXiv
Apriel-Nemotron-15B-Thinker
Shruthan Radhakrishna, Soham Parikh, Gopal Sarda, Anil Turkkan, and 31 more authors
@misc{radhakrishna2025aprielnemotron15bthinker,title={Apriel-Nemotron-15B-Thinker},author={Radhakrishna, Shruthan and Parikh, Soham and Sarda, Gopal and Turkkan, Anil and Vohra, Quaizar and Li, Raymond and Jhamb, Dhruv and Ogueji, Kelechi and Shukla, Aanjaneya and Bamgbose, Oluwanifemi and Liang, Toby and Kumar, Luke and Ostapenko, Oleksiy and Malay, Shiva Krishna Reddy and Tiwari, Aman and Bogavelli, Tara and Yadav, Vikas and Mehta, Jash and Mittal, Saloni and Kalkunte, Akshay and Pattnaik, Pulkit and Slimi, Khalil and Sreeram, Anirudh and Nair, Jishnu and Oladipo, Akintunde and Maiya, Shashank and Mahajan, Khyati and Maheshwary, Rishabh and Hashemi, Masoud and Mudumba, Sai Rajeswar and Madhusudhan, Sathwik Tejaswi and Scholak, Torsten and Paquet, Sebastien and Davasam, Sagar and Sunkara, Srinivas},year={2025},eprint={2508.10948},archiveprefix={arXiv},primaryclass={cs.LG},}
Patent
Platform Agnostic Scalable and High-Performance Semantic Search Framework
Sriram Puttagunta, Jishnu Sethumadhavan Nair, Bidyapati Pradhan, Nirali Dineshbhai Popat, and 4 more authors
Sep 2025
US Patent 12,430,374 B1 (Issued: September 30, 2025)
@misc{patent_semantic_search2025,title={Platform Agnostic Scalable and High-Performance Semantic Search Framework},author={Puttagunta, Sriram and Sethumadhavan Nair, Jishnu and Pradhan, Bidyapati and Popat, Nirali Dineshbhai and Ramachandran, Sravan and Mittal, Vipul and Subramanian, Seganrasan and Chenna, Ranga Prasad},year={2025},month=sep,day={30},note={US Patent 12,430,374 B1 (Issued: September 30, 2025)},publisher={ServiceNow Inc},}
Patent
Framework Neutral and Updatable Clustering Model
Sriram Puttagunta, Jishnu Sethumadhavan Nair, Bidyapati Pradhan, Nirali Dineshbhai Popat, and 4 more authors
@misc{patent_clustering2025,title={Framework Neutral and Updatable Clustering Model},author={Puttagunta, Sriram and Sethumadhavan Nair, Jishnu and Pradhan, Bidyapati and Popat, Nirali Dineshbhai and Ramachandran, Sravan and Mittal, Vipul and Subramanian, Seganrasan and Chenna, Ranga Prasad},year={2025},month=jul,note={US Patent Application 20250245244 A1 (Pending)},publisher={ServiceNow Inc},}