Mehmet Onur Keskin

Amsterdam, the Netherlands

I am an AI researcher developing intelligent systems that model (users’ or institutions’) needs and preferences while trying to find and persuade mutually optimal solutions in highly uncertain, complex, and mostly real-time environments.

I earned an M.Sc. in Artificial Intelligence (summa cum laude) under the supervision of Dr. Reyhan Aydogan at the Interactive Intelligent Systems Lab, Ozyegin University. There, I investigated how autonomous agents can model and adapt to human preferences, arguments, and nonverbal affective signals during real-time interactions. This work was published in AI conferences (e.g., AAMAS, IJCAI) and journals (JAAMAS, IEEE THMS, ACM TiiS).

My academic nomad journey significantly changed my perspective on impact-oriented research through several international visiting research positions:

At TU Delft’s Interactive Intelligence Group, under Prof. Catholijn Jonker, I developed multimodal frameworks for human-agent negotiation and led large-scale experiments to address the challenges of modelling complex, real-time human signals.
At the University of Cambridge’s Affective Intelligence & Robotics Lab, under the supervision of Prof. Hatice Gunes, I applied continual learning to a personalized agent strategy. The goal was to convert arousal and valence feedback from users’ facial expressions into instant decision-making functions for human-robot negotiation.
At Brown University’s E-GLAMOR Group, led by Prof. Amy Greenwald, we developed a supply chain management agent. The agent optimizes for long-term objectives (e.g, maximizing profit and partnership trust while minimizing market volatility risks). I introduced a combinatorial dispatching algorithm rooted in computational game theory. This enables agents to align short-term market-making actions with long-term objectives. We were runner-up in the supply chain management league at IJCAI 2021 (SCML 2021).
At the Collaborative Robotics and Intelligent Systems Institute of Oregon State University (NSF REU Scholar), I focused on optimization algorithms for 3D printing under the guidance of Prof. Yigit Menguc.

During my academic career, I have served as a Program Committee Member at AAMAS. I took leadership roles in the IEEE CS Turkey Chapter. I organized the Automated Negotiating Agents Competition at AAMAS 2023 and regularly competed in agent competitions. Notably, I won the IBM Watson’s Human Multi-Agent Immersive Negotiation Challenge at IJCAI 2020. My work has received recognition from the IEEE CS Richard Merwin Student Scholarship, Ozyegin University’s Best Research Assistant Award, and the Heidelberg Laureate Forum (HLF) as a young researcher.

I have always been motivated to deploy multi-agent systems and reasoning architectures to real-life challenges. I co-founded Kanzy Financial Technologies in London, UK, and served as CTO. We developed Finguru, an all-in-one brokerage app for self-directed investors looking for a personalized wealth-building experience with AI assistance. To develop an AI agent providing trustworthy guidance within strict compliance constraints, my core focus was on architecting the scalable agent orchestration, persistent memory for personalization, and user preference and risk modelling.

This applied start-up venture received support from NVIDIA Inception and Google for Startups AI, as well as securing pre-seed round funding. But I am currently seeking a PhD position to advance foundational research on adaptive, reliable, and cooperative multi-agent systems, rather than focusing on strict regulations.

Recent News

Nov 2025	Currently on leave from my role as Co-founder & CTO at Kanzy Financial Technologies to focus on exploring foundational AI research opportunities.
Oct 2025	Our paper “An Adaptive Emotion-Aware Strategy for Human-Agent Negotiation: Insights from Real-World Human-Robot Experiments” was presented at the 25th ACM International Conference on Intelligent Virtual Agents (IVA).
Jul 2025	Our paper “Evaluating Modern RAG: Textual, Multimodal, Dense, and Late Interaction Pipelines” was presented at the Workshop on Information Retrieval’s Role in RAG Systems at the 48th ACM SIGIR Conference.
May 2025	Kanzy Financial Technologies was accepted into the NVIDIA Inception and the Google for Startups AI programs, recognizing our work in applying AI to financial services.
May 2025	Honored to be selected to attend the prestigious Heidelberg Laureate Forum (HLF) as a Young Researcher, connecting with laureates in Mathematics and Computer Science.

Selected Publications

IJCAI 2024
NEGOTIATOR: a comprehensive framework for human-agent negotiation integrating preferences, interaction, and emotion

Mehmet Onur Keskin, Berk Buzcu, Berkecan Koçyiğit, and 3 more authors

In Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea

Aug 2024

Abs Bib DOI PDF

The paper introduces a comprehensive human-agent negotiation framework designed to facilitate the development and evaluation of research studies on human-agent negotiation without building each component from scratch. Leveraging the interoperability and reusability of its components, this framework offers various functionalities, including speech-to-text conversion, emotion recognition, a repository of negotiation strategies, and an interaction manager capable of managing gestures designed for Nao, Pepper, and QT, and coordinating message exchanges in a turn-taking fashion. This framework aims to lower the entry barrier for researchers in human-agent negotiation by providing a versatile platform that supports a wide range of research directions, including affective computing, natural language processing, decision-making, and non-verbal communication.
```
@inproceedings{10.24963/ijcai.2024/1012,
  month = aug,
  booktitle = {Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea},
  pid = {12},
  author = {Keskin, Mehmet Onur and Buzcu, Berk and Ko\c{c}yi\u{g}it, Berkecan and \c{C}akan, Umut and Do\u{g}ru, Anil and Aydo\u{g}an, Reyhan},
  title = {NEGOTIATOR: a comprehensive framework for human-agent negotiation integrating preferences, interaction, and emotion},
  year = {2024},
  isbn = {978-1-956792-04-1},
  url = {https://doi.org/10.24963/ijcai.2024/1012},
  doi = {10.24963/ijcai.2024/1012},
  articleno = {1012},
  numpages = {4},
  series = {IJCAI '24}
}
```
IJCAI 2024
NegoLog: An Integrated Python-based Automated Negotiation Framework with Enhanced Assessment Components

Anıl Doğru, Mehmet Onur Keskin, Catholijn M. Jonker, and 2 more authors

In Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea

Aug 2024

Abs Bib DOI PDF

The complexity of automated negotiation research calls for dedicated, user-friendly research frameworks that facilitate advanced analytics, comprehensive loggers, visualization tools, and auto-generated domains and preference profiles. This paper introduces NegoLog, a platform that provides advanced and customizable analysis modules to agent developers for exhaustive performance evaluation. NegoLog introduces an automated scenario and tournament generation tool in its Web-based user interface so that the agent developers can adjust the competitiveness and complexity of the negotiations. One of the key novelties of the NegoLog is an individual assessment of preference estimation models independent of the strategies.
```
@inproceedings{ijcai2024p998,
  month = aug,
  booktitle = {Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea},
  pid = {11},
  title = {NegoLog: An Integrated Python-based Automated Negotiation Framework with Enhanced Assessment Components},
  author = {Doğru, Anıl and Keskin, Mehmet Onur and Jonker, Catholijn M. and Baarslag, Tim and Aydoğan, Reyhan},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor = {Larson, Kate},
  pages = {8640--8643},
  year = {2024},
  doi = {10.24963/ijcai.2024/998},
  url = {https://doi.org/10.24963/ijcai.2024/998}
}
```
JAAMAS 2024
Decentralized multi-agent path finding framework and strategies based on automated negotiation

Mehmet Onur Keskin, Furkan Cantürk, Cihan Eran, and 1 more author

Autonomous Agents and Multi-Agent Systems

Mar 2024

Abs Bib DOI PDF

This paper introduces a negotiation framework to solve the Multi-Agent Path Finding (MAPF) Problem for self-interested agents in a decentralized fashion. The framework aims to achieve a good trade-off between the privacy of the agents and the effectiveness of solutions. Accordingly, a token-based bilateral negotiation protocol and two negotiation strategies are presented. The experimental results over four different settings of the MAPF problem show that the proposed approach could find conflict-free path solutions albeit suboptimally, especially when the search space is large and high-density. In contrast, Explicit Estimation Conflict-Based Search (EECBS) struggles to find optimal solutions. Besides, deploying a sophisticated negotiation strategy that utilizes information about local density for generating alternative paths can yield remarkably better solution performance in this negotiation framework.
```
@article{Keskin2024_MAPF,
  month = mar,
  journal = {Autonomous Agents and Multi-Agent Systems},
  pid = {08},
  author = {Keskin, Mehmet Onur and Cant{\"{u}}rk, Furkan and Eran, Cihan and Aydo{\u{g}}an, Reyhan},
  title = {Decentralized multi-agent path finding framework and strategies based on automated negotiation},
  year = {2024},
  doi = {10.1007/s10458-024-09639-8},
  url = {https://doi.org/10.1007/s10458-024-09639-8}
}
```
IEEE THMS 2022
Would You Imagine Yourself Negotiating With a Robot, Jennifer? Why Not?

Reyhan Aydoğan, Mehmet Onur Keskin, and Umut Çakan

IEEE Transactions on Human-Machine Systems

Feb 2022

Abs Bib DOI PDF

With the improvement of intelligent systems and robotics, social robots are becoming part of our society. To accomplish complex tasks, robots and humans may need to collaborate, and when necessary, they need to negotiate with each other. While designing such socially interacting robots, it is crucial to consider human factors such as facial expression, emotions, and body language. Since gestures play a crucial role in interaction, this article studies the effect of gestures in human–robot negotiation experiments. Additionally, it compares the performance of variants of the well-known negotiation tactics (i.e., time-based and behavior-based) in automated negotiation literature in the context of human–robot negotiations. Our experimental results support the finding in automated negotiation. That is, the robot gained higher utility when it imitates its opponent’s bidding strategy than employing a time-based negotiation strategy. When adopting a behavior-based technique, there is a statistically significant effect of gestures on the underlying negotiation process, and, therefore, on negotiation outcome.
```
@article{9609976,
  month = feb,
  journal = {IEEE Transactions on Human-Machine Systems},
  pid = {05},
  author = {Aydoğan, Reyhan and Keskin, Mehmet Onur and Çakan, Umut},
  title = {Would You Imagine Yourself Negotiating With a Robot, Jennifer? Why Not?},
  year = {2022},
  volume = {52},
  number = {1},
  pages = {41-51},
  doi = {10.1109/THMS.2021.3121664}
}
```
AAMAS 2021
Solver Agent: Towards Emotional and Opponent-Aware Agent for Human-Robot Negotiation

Mehmet Onur Keskin, Umut Çakan, and Reyhan Aydogan

In International Conference on Autonomous Agents and Multiagent Systems

Jan 2021

Abs Bib DOI PDF

Negotiation is one of the crucial processes for resolving conflicts between parties. In automated negotiation, agent designers mostly take opponent’s offers and the remaining time into account while designing their strategies. While designing a negotiating agent interacting with a human directly, other information such as opponent’s emotional changes during the negotiation can establish a better interaction and reach an admissible settlement for joint interests. Accordingly, this paper proposes a bidding strategy for humanoid robots, which incorporates their opponents’ emotional states and awareness of the agent’s changing behavior.
```
@inproceedings{inproceedings,
  month = jan,
  booktitle = {International Conference on Autonomous Agents and Multiagent Systems},
  doi = {10.5555/3463952.3464158},
  pid = {02},
  author = {Keskin, Mehmet Onur and Çakan, Umut and Aydogan, Reyhan},
  year = {2021},
  pages = {},
  title = {Solver Agent: Towards Emotional and Opponent-Aware Agent for Human-Robot Negotiation}
}
```