Vinod Muthusamy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Vinod Muthusamy has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Aleksander Slominski
IBM Research
4 shared publications - Vatche Isahagian
IBM Research AI
4 shared publications - Hans-Arno Jacobsen
University of Toronto, Toronto, Canada
2 shared publications - K. R. Jayaram
IBM Research AI Yorktown Heights, USA
2 shared publications - Praveen Venkateswaran
IBM Research AI Yorktown Heights, USA
2 shared publications Vatche Ishakian
Bentley University
2 shared publications
Latest Publications
- Open Access
- 26 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 1340-1344
Agent Lifecycle Toolkit (ALTK): Reusable Middleware Components for Robust AI Agents
As AI agents move from demos into enterprise deployments, their failure modes become consequential: a misinterpreted tool argument can corrupt production data, a silent reasoning error can go undetected until damage is done, and outputs that violate organizational policy can create legal or compliance risk. Yet, most agent frameworks leave builders to handle these failure modes ad hoc, resulting in brittle, one-off safeguards that are hard to reuse or maintain. We present the Agent Lifecycle Toolkit (ALTK), an open-source collection of modular middleware components that systematically address these gaps across the full agent lifecycle. Across the agent lifecycle, we identify opportunities to intervene and improve, namely, post‑user-request, pre‑LLM prompt conditioning, post‑LLM output processing, pre‑tool validation, post‑tool result checking, and pre‑response assembly. ALTK provides modular middleware that detects, repairs, and mitigates common failure modes. It offers consistent interfaces that fit naturally into existing pipelines. It is compatible with low-code and no-code tools such as the ContextForge MCP Gateway and Langflow. Finally, it significantly reduces the effort of building reliable, production‑grade agents.
- 26 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 1340-1344
Open AccessAgent Lifecycle Toolkit (ALTK): Reusable Middleware Components for Robust AI Agents
As AI agents move from demos into enterprise deployments, their failure modes become consequential: a misinterpreted tool argument can corrupt production data, a silent reasoning error can go undetected until damage is done, and outputs that violate organizational policy can create legal or compliance risk. Yet, most agent frameworks leave builders to handle these failure modes ad hoc, resulting in brittle, one-off safeguards that are hard to reuse or maintain. We present the Agent Lifecycle Toolkit (ALTK), an open-source collection of modular middleware components that systematically address these gaps across the full agent lifecycle. Across the agent lifecycle, we identify opportunities to intervene and improve, namely, post‑user-request, pre‑LLM prompt conditioning, post‑LLM output processing, pre‑tool validation, post‑tool result checking, and pre‑response assembly. ALTK provides modular middleware that detects, repairs, and mitigates common failure modes. It offers consistent interfaces that fit naturally into existing pipelines. It is compatible with low-code and no-code tools such as the ContextForge MCP Gateway and Langflow. Finally, it significantly reduces the effort of building reliable, production‑grade agents.
- Open Access
- 15 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 7-9
In-Context Example Ordering for LLM-Based API Sequence Generation
- 15 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 7-9
Open AccessIn-Context Example Ordering for LLM-Based API Sequence Generation
- Open Access
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
A Conversational Assistant Framework for Automation
This paper describes the design and implementation of an AI-enabled conversational assistant to enhance the efficacy of Business Process Automation (BPA) tools in the context of IBM's watsonx Orchestrate platform. It leverages the capabilities of large language models (LLMs) to interpret natural language instructions and unstructured documents, thereby simplifying the workflow development process. We detail the assistant framework's architecture, which supports extendable, event-driven interactions and dynamically integrates multiple agents. We demonstrate its utility through a real sales prospecting use case, which showcases its ability to facilitate the authoring of workflows, decision rules, and data models. Additionally, our empirical evaluation demonstrates the accuracy of routing user requests, efficacy in workflow authoring, and the precision of API field mapping recommendations.
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessA Conversational Assistant Framework for Automation
This paper describes the design and implementation of an AI-enabled conversational assistant to enhance the efficacy of Business Process Automation (BPA) tools in the context of IBM's watsonx Orchestrate platform. It leverages the capabilities of large language models (LLMs) to interpret natural language instructions and unstructured documents, thereby simplifying the workflow development process. We detail the assistant framework's architecture, which supports extendable, event-driven interactions and dynamically integrates multiple agents. We demonstrate its utility through a real sales prospecting use case, which showcases its ability to facilitate the authoring of workflows, decision rules, and data models. Additionally, our empirical evaluation demonstrates the accuracy of routing user requests, efficacy in workflow authoring, and the precision of API field mapping recommendations.
- Open Access
- 11 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 29-30
Publish-subscribe with large language models
Large Language Models (LLMs) have raised the expectations of users to be able to use natural language (NL) to interact with computer systems such as publish/subscribe systems. We introduce a design that extends existing pub/sub paradigms by adding support for NL queries to subscribe to publications of unstructured text messages. To make the system easier for users our system can also customize the notifications, using LLMs to generate targeted NL answers from the matching publications.
- 11 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 29-30
Open AccessPublish-subscribe with large language models
Large Language Models (LLMs) have raised the expectations of users to be able to use natural language (NL) to interact with computer systems such as publish/subscribe systems. We introduce a design that extends existing pub/sub paradigms by adding support for NL queries to subscribe to publications of unstructured text messages. To make the system easier for users our system can also customize the notifications, using LLMs to generate targeted NL answers from the matching publications.
- Open Access
Adaptive Aggregation For Federated Learning
Advances in federated learning (FL) algorithms,along with technologies like differential privacy and homomorphic encryption, have led to FL being increasingly adopted and used in many application domains. This increasing adoption has led to rapid growth in the number, size (number of participants/parties) and diversity (intermittent vs. active parties) of FL jobs. Many existing FL systems, based on centralized (often single) model aggregators are unable to scale to handle large FL jobs and adapt to parties' behavior. In this paper, we present a new scalable and adaptive architecture for FL aggregation. First, we demonstrate how traditional tree overlay based aggregation techniques (from P2P, publish-subscribe and stream processing research) can help FL aggregation scale, but are ineffective from a resource utilization and cost standpoint. Next, we present the design and implementation of AdaFed, which uses serverless/cloud functions to adaptively scale aggregation in a resource efficient and fault tolerant manner. We describe how AdaFed enables FL aggregation to be dynamically deployed only when necessary, elastically scaled to handle participant joins/leaves and is fault tolerant with minimal effort required on the (aggregation) programmer side. We also demonstrate that our prototype based on Ray scales to thousands of participants, and is able to achieve a >90% reduction in resource requirements and cost, with minimal impact on aggregation latency.
Open AccessAdaptive Aggregation For Federated Learning
Advances in federated learning (FL) algorithms,along with technologies like differential privacy and homomorphic encryption, have led to FL being increasingly adopted and used in many application domains. This increasing adoption has led to rapid growth in the number, size (number of participants/parties) and diversity (intermittent vs. active parties) of FL jobs. Many existing FL systems, based on centralized (often single) model aggregators are unable to scale to handle large FL jobs and adapt to parties' behavior. In this paper, we present a new scalable and adaptive architecture for FL aggregation. First, we demonstrate how traditional tree overlay based aggregation techniques (from P2P, publish-subscribe and stream processing research) can help FL aggregation scale, but are ineffective from a resource utilization and cost standpoint. Next, we present the design and implementation of AdaFed, which uses serverless/cloud functions to adaptively scale aggregation in a resource efficient and fault tolerant manner. We describe how AdaFed enables FL aggregation to be dynamically deployed only when necessary, elastically scaled to handle participant joins/leaves and is fault tolerant with minimal effort required on the (aggregation) programmer side. We also demonstrate that our prototype based on Ray scales to thousands of participants, and is able to achieve a >90% reduction in resource requirements and cost, with minimal impact on aggregation latency.
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 466-468
Summarization of Multidimensional Process Traces for Analysis under Edit-distance Constraints
Motivation: Business processes and workflows exhibit data models in the form of multi-dimensional sequence of objects. As exemplified in Figure 1 , a business processes, represented as a directed acyclic graph of activities, generates execution traces as a sequence of activities with associated multi-dimensional attributes. For example, an activity in the loan application process can contain information about the person and the department that are responsible for the activity, the person who performs the activity, and the group to which she belongs. Businesses analyze operational data for insights such as understanding workflow patterns and bottlenecks [1] , verifying conformance to policies or regulations [2] , or revealing clusters of common behavior [3] .
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 466-468
Summarization of Multidimensional Process Traces for Analysis under Edit-distance Constraints
Motivation: Business processes and workflows exhibit data models in the form of multi-dimensional sequence of objects. As exemplified in Figure 1 , a business processes, represented as a directed acyclic graph of activities, generates execution traces as a sequence of activities with associated multi-dimensional attributes. For example, an activity in the loan application process can contain information about the person and the department that are responsible for the activity, the person who performs the activity, and the group to which she belongs. Businesses analyze operational data for insights such as understanding workflow patterns and bottlenecks [1] , verifying conformance to policies or regulations [2] , or revealing clusters of common behavior [3] .
- Open Access
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 108-109
Towards Enterprise-Ready AI Deployments Minimizing the Risk of Consuming AI Models in Business Applications
The stochastic nature of artificial intelligence (AI) models introduces risk to business applications that use AI models without careful consideration. This paper offers an approach to use AI techniques to gain insights on the usage of the AI models and control how they are deployed to a production application.
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 108-109
Open AccessTowards Enterprise-Ready AI Deployments Minimizing the Risk of Consuming AI Models in Business Applications
The stochastic nature of artificial intelligence (AI) models introduces risk to business applications that use AI models without careful consideration. This paper offers an approach to use AI techniques to gain insights on the usage of the AI models and control how they are deployed to a production application.
- Open Access
- 13 June 2016
- Published by Association for Computing Machinery (ACM)
- p. 199-204
Lessons learned using a process mining approach to analyze events from distributed applications
The execution of distributed applications are captured by the events generated by the individual components. However, understanding the behavior of these applications from their event logs can be a complex and error prone task, compounded by the fact that applications continuously change rendering any knowledge obsolete. We describe our experiences applying a suite of process-aware analytic tools to a number of real world scenarios, and distill our lessons learned. For example, we have seen that these tools are used iteratively, where insights gained at one stage inform the configuration decisions made at an earlier stage. As well, we have observed that data onboarding, where the raw data is cleaned and transformed, is the most critical stage in the pipeline and requires the most manual effort and domain knowledge. In particular, missing, inconsistent, and low-resolution event time stamps are recurring problems that require better solutions. The experiences and insights presented here will assist practitioners applying process analytic tools to real scenarios, and reveal to researchers some of the more pressing challenges in this space.
- 13 June 2016
- Published by Association for Computing Machinery (ACM)
- p. 199-204
Open AccessLessons learned using a process mining approach to analyze events from distributed applications
The execution of distributed applications are captured by the events generated by the individual components. However, understanding the behavior of these applications from their event logs can be a complex and error prone task, compounded by the fact that applications continuously change rendering any knowledge obsolete. We describe our experiences applying a suite of process-aware analytic tools to a number of real world scenarios, and distill our lessons learned. For example, we have seen that these tools are used iteratively, where insights gained at one stage inform the configuration decisions made at an earlier stage. As well, we have observed that data onboarding, where the raw data is cleaned and transformed, is the most critical stage in the pipeline and requires the most manual effort and domain knowledge. In particular, missing, inconsistent, and low-resolution event time stamps are recurring problems that require better solutions. The experiences and insights presented here will assist practitioners applying process analytic tools to real scenarios, and reveal to researchers some of the more pressing challenges in this space.
- 27 November 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 22 (5) , 1516-1530
Infrastructure-Free Content-Based Publish/Subscribe
Peer-to-peer (P2P) networks can offer benefits to distributed content-based publish/subscribe data dissemination systems. In particular, since a P2P network's aggregate resources grow as the number of participants increases, scalability can be achieved using no infrastructure other than the participants' own resources. This paper proposes algorithms for supporting content-based publish/subscribe in which subscriptions can specify a range of interest and publications a range of values. The algorithms are built over a distributed hash table abstraction and are completely decentralized. Load balance is addressed by subscription delegation away from overloaded peers and a bottom-up tree search technique that avoids root hotspots. Furthermore, fault tolerance is achieved with a lightweight replication scheme that quickly detects and recovers from faults. Experimental results support the scalability and fault-tolerance properties of the algorithms: For example, doubling the number of subscriptions does not double internal system messages, and even the simultaneous failure of 20% of the peers in the system requires less than 2 min to fully recover.
- 27 November 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 22 (5) , 1516-1530
Infrastructure-Free Content-Based Publish/Subscribe
Peer-to-peer (P2P) networks can offer benefits to distributed content-based publish/subscribe data dissemination systems. In particular, since a P2P network's aggregate resources grow as the number of participants increases, scalability can be achieved using no infrastructure other than the participants' own resources. This paper proposes algorithms for supporting content-based publish/subscribe in which subscriptions can specify a range of interest and publications a range of values. The algorithms are built over a distributed hash table abstraction and are completely decentralized. Load balance is addressed by subscription delegation away from overloaded peers and a bottom-up tree search technique that avoids root hotspots. Furthermore, fault tolerance is achieved with a lightweight replication scheme that quickly detects and recovers from faults. Experimental results support the scalability and fault-tolerance properties of the algorithms: For example, doubling the number of subscriptions does not double internal system messages, and even the simultaneous failure of 20% of the peers in the system requires less than 2 min to fully recover.
- Open Access
- 1 December 2008
- Published by Association for Computing Machinery (ACM)
- p. 31-36
SLA-driven distributed application development
- 1 December 2008
- Published by Association for Computing Machinery (ACM)
- p. 31-36
Open AccessSLA-driven distributed application development