SOUVA MAJUMDER
Headshot
Bio
Souva Majumder is the Director of AI Technologies at Apollonius Computational Business Solutions and the Chair of IEEE SA IC 25-003. An alumnus of IIT Kharagpur in Industrial & Systems Engineering, he leads innovation in Generative and Agentic AI. His key interest lies in the evaluation of multi-agentic systems, and he holds a software patent on Evaluation of Multi-Agentic Systems using Agent-Agent as a Judge. He specializes in multi-agent architectures, RAG pipelines, vector databases, and advanced prompt engineering. Souva has provided corporate training on Agentic AI for Business Transformation at Infosys, Morningstar India, IIM Lucknow, and other leading institutions.
For those giving lightening talks, are you able to share your slides in a ‘Google Slides link’ to add to our Master Google Slides deck?
-Optional Questions if you’re representing a Standards Body -
1 Paragraph description of your protocol/standard?
There exists a broad class of complex, previously unsolved problems—spanning fields such as medicine, climate science, engineering, and the natural sciences—for which viable solutions are not yet available in the public domain. These problems are inherently non-trivial and often fall outside the knowledge distribution on which standard large language models (LLMs) are pretrained. As such, current LLMs are ill-suited to address these challenges directly, due to both their epistemic limitations and lack of domain-specific reasoning.
To address this, we propose evaluation of a customized multi-agent system capable of collaboratively reasoning about and attempting solutions to such novel problems. The primary objective is to explore whether a system of coordinated Artificial Intelligence (AI) agents—each with distinct roles, tools, and knowledge access—can engage in goal-directed problem-solving in uncharted domains. While we anticipate that such a system may generate interesting hypotheses or partial insights, it is also likely to produce hallucinations, factual inaccuracies, and logical errors, especially when operating outside familiar data distributions.
To address this, we propose evaluation of a customized multi-agent system capable of collaboratively reasoning about and attempting solutions to such novel problems. The primary objective is to explore whether a system of coordinated Artificial Intelligence (AI) agents—each with distinct roles, tools, and knowledge access—can engage in goal-directed problem-solving in uncharted domains. While we anticipate that such a system may generate interesting hypotheses or partial insights, it is also likely to produce hallucinations, factual inaccuracies, and logical errors, especially when operating outside familiar data distributions.
Bio (Short)
Current needs for your protocol/standard?
We need working group participants who is willing to develop the Software
Email contact for people interested in your protocol/standard (this will be shared publicly on the notion page)
aiagent-officers@ieee.org
Link for participants to learn more about the protocol/standard:
LinkedIn
https://in.linkedin.com/in/avik123
Organization
Apollonius Computational Business Solutions OPC Pvt. Ltd
Protocol / Standard
IEEE SA Industry Connected Program IC 25 - 003
Where is the protocol/standard being housed? (1)
IEEE
X
No
logo of your company or protocol/standard - please make sure you give us the file in
Made with Bullet