---
title: "Agentic AI for Human Resources — paper by Yuksel et al. (2026)"
description: "This system demonstration from aiXplain presents HR Manager, a modular pipeline of LLM agents for candidate assessment… 17 references."
authors: ["Kamer Ali Yuksel", "Abdul Basit Anees", "Ashraf Hatim Elneima", "Sanjika Hewavitharana", "Mohamed Al-Badrashiny", "Hassan Sawaf"]
year: 2026
venue: "Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)"
source: "https://aclanthology.org/2026.eacl-demo.24/"
doi: "10.18653/v1/2026.eacl-demo.24"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/yuksel-2026-agentic-ai-human-resources-llm/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yuksel-2026-agentic-ai-human-resources-llm/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/21/) / Agentic AI for Human Resources: LLM-Driven Candidate Assessment

# Agentic AI for Human Resources: LLM-Driven Candidate Assessment

**Authors:** Kamer Ali Yuksel, Abdul Basit Anees, [Ashraf Hatim Elneima](https://md.research.elamur.ai/free/en/authors/ashraf-hatim-elneima/), Sanjika Hewavitharana, [Mohamed Al-Badrashiny](https://md.research.elamur.ai/free/en/authors/mohamed-al-badrashiny/) and [Hassan Sawaf](https://md.research.elamur.ai/free/en/authors/hassan-sawaf/) · **Year:** 2026 · **Venue:** [Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)](https://md.research.elamur.ai/free/en/venues/eacl/) ([EACL 2026](https://md.research.elamur.ai/free/en/venues/eacl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.eacl-demo.24/) · [PDF](https://aclanthology.org/2026.eacl-demo.24.pdf) · [DOI 10.18653/v1/2026.eacl-demo.24](https://doi.org/10.18653/v1/2026.eacl-demo.24) · [OpenAlex](https://openalex.org/W7140106877)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Evaluating job applicants is expensive and error-prone. Keyword-based applicant tracking tools miss nuance, and human reviewers suffer from fatigue, disagreement and bias. This system demonstration from aiXplain presents HR Manager, a modular pipeline of LLM agents for candidate assessment:

- **Rubric creation:** one agent converts a job title and description into a role-tailored rubric.
- **Interviews:** another agent writes video interview questions.
- **Assessment:** a third agent merges CVs, interview transcripts, HR notes and chat logs into markdown reports. Each report gives Low/Medium/High ratings with evidence-backed justifications.
- **Comparison and ranking:** further agents format, compare and rank applicants.

For ranking, the authors reuse their earlier financial-asset technique. The LLM orders small candidate groups ("mini-tournaments"). A Plackett–Luce model then combines these permutations into global utility scores, and active learning selects the most informative groups to query next.

On a real ranking task scored against human expert orderings, peak NDCG@K ranged from about 0.51 to 0.57. The ranking stabilized within roughly 20 iterations. The main contribution is an interpretable, auditable decision-support design for hiring. However, the available text reports no baseline comparison or dataset size.

## Key points
- The Criteria Generation Agent outputs a YAML rubric, typically with 12–20 dimensions such as Industry-Specific Fit, Functional Fit, Motivation, Drive and Continuous Learning. Each dimension defines what every proficiency level means.
- The system has these agents: Criteria Generation, Video Question Generation, Assessment Generator, Formatter, Comparison and Ranking. It also includes a video-analysis module and a Feedback Integration Module that links assessments to post-hire outcomes. The conclusion, however, presents the outcome feedback loop as future work.
- Peak NDCG@K against the human ranking: 0.5134 (10%), 0.5455 (15%), 0.5655 (20%) and 0.5703 (25%), measured over 30 active-learning iterations.
- Convergence was tracked with two signals:
  - Kendall-τ between successive iterations, which rose quickly early on.
  - The norm of the utility-vector change (Δu), which fell sharply after iteration 10.
- Stated limitations include:
  - sensitivity to poor job descriptions or unclear candidate materials;
  - non-verbal analysis currently limited to facial expressions, which conflicts with the architecture section's claims about tone and body language;
  - risks of rubric drift and of bias amplification across active-learning rounds.
- A live demo is available at hrmanager.aixplain.com.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (4): [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🤝 Safety Research](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · 626 works · [💼 Business, Management and Accounting](https://md.research.elamur.ai/free/en/shelves/business-management-and-accounting/) · 40 works · [💼 Organizational Behavior and Human Resource Management](https://md.research.elamur.ai/free/en/shelves/business-management-and-accounting/organizational-behavior-and-human-resource-management/) · 15 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works

## Cite

Cite as: Kamer Ali Yuksel, Abdul Basit Anees, Ashraf Hatim Elneima, Sanjika Hewavitharana, Mohamed Al-Badrashiny and Hassan Sawaf. 2026. Agentic AI for Human Resources: LLM-Driven Candidate Assessment. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{yuksel2026agentic,
  title = {{Agentic AI for Human Resources: LLM-Driven Candidate Assessment}},
  author = {Kamer Ali Yuksel, Abdul Basit Anees, Ashraf Hatim Elneima, Sanjika Hewavitharana, Mohamed Al-Badrashiny and Hassan Sawaf},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)},
  year = {2026},
  doi = {10.18653/v1/2026.eacl-demo.24},
  url = {https://doi.org/10.18653/v1/2026.eacl-demo.24}
}
```

- [← previous AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments](https://md.research.elamur.ai/free/en/papers/xi-2026-agentgym2-benchmarking-large-language-model/)
- [next → Agentic CLEAR: Automating Multi-Level Evaluation of LLM Agents](https://md.research.elamur.ai/free/en/papers/yehudai-2026-agentic-clear-automating-multi-level/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/21/)

## References

17 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Maximilian Balandat, Brian Karrer, Daniel R. Jiang, Samuel Daulton, Benjamin Letham, Andrew Gordon Wilson, and Eytan Bakshy. 2020. Botorch: A framework for efficient monte-carlo bayesian optimization. \#\# Ethics Statement This system aims to democratize access to fair candidate assessments, reducing… [find](https://scholar.google.com/scholar?q=Maximilian+Balandat%2C+Brian+Karrer%2C+Daniel+R%2E+Jiang%2C+Samuel+Daulton%2C+Benjamin+Letham%2C+Andrew+Gordon+Wilson%2C+and+Eytan+Bakshy%2E+2020%2E+Botorch%3A+A+framework+for+efficient+monte%2Dcarlo+bayesian+optimization%2E)
2. R. V. K. Bevara, N. R. Mannuru, S. P. Karedla, B. Lund, T. Xiao, H. Pasem, S. C. Dronavalli, and S. Rudreshkumar. 2025. Resume2vec: Transforming applicant tracking systems with intelligent resume embeddings for precise candidate matching. Electronics, 14(4):794. [find](https://scholar.google.com/scholar?q=R%2E+V%2E+K%2E+Bevara%2C+N%2E+R%2E+Mannuru%2C+S%2E+P%2E+Karedla%2C+B%2E+Lund%2C+T%2E+Xiao%2C+H%2E+Pasem%2C+S%2E+C%2E+Dronavalli%2C+and+S%2E+Rudreshkumar%2E+2025%2E+Resume2vec%3A+Transforming+applicant+tracking+systems+with+intelligent+resume+embe)
3. Aurélien Garivier and Olivier Cappé. 2011. The kl-ucb algorithm for bounded stochastic bandits and beyond. In Proceedings of the 24th Annual Conference on Learning Theory (COLT 2011), volume 19 of Proceedings of Machine Learning Research, pages 359–376. [find](https://scholar.google.com/scholar?q=Aur%C3%A9lien+Garivier+and+Olivier+Capp%C3%A9%2E+2011%2E+The+kl%2Ducb+algorithm+for+bounded+stochastic+bandits+and+beyond%2E+In+Proceedings+of+the+24th+Annual+Conference+on+Learning+Theory+%28COLT+2011%29%2C+volume+19+of+Pro)
4. Preetam Ghosh and Vaishali Sadaphal. 2023. Jo-brecogpt – explainable job recommendations using llms. [find](https://scholar.google.com/scholar?q=Preetam+Ghosh+and+Vaishali+Sadaphal%2E+2023%2E+Jo%2Dbrecogpt+%E2%80%93+explainable+job+recommendations+using+llms%2E)
5. Taicheng Guo, Xiuying Chen, Yaqi Wang, Ruidi Chang, Shichao Pei, Nitesh V. Chawla, Olaf Wiest, and Xiangliang Zhang. 2024. [Large language model based multi-agents: A survey of progress and challenges](https://md.research.elamur.ai/free/en/papers/guo-2024-large-language-model-based/).
6. Darsh Kanikar, Pratyush Jain, Siddhi Jain, and Lalit Purohit. 2025. Systematic review of methods for analysis of resumes. In Proceedings of the International Conference on Computer Science and Communication Engineering, pages 301–308. Atlantis Press. Systematic survey of resume analyzers, including… [find](https://scholar.google.com/scholar?q=Darsh+Kanikar%2C+Pratyush+Jain%2C+Siddhi+Jain%2C+and+Lalit+Purohit%2E+2025%2E+Systematic+review+of+methods+for+analysis+of+resumes%2E+In+Proceedings+of+the+International+Conference+on+Computer+Science+and+Communi)
7. Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2023. [Large language models are zero-shot reasoners](https://md.research.elamur.ai/free/en/papers/kojima-2022-large-language-models-zero-shot/).
8. Lucy Liang and Kristen Grauman. 2014. Beyond comparing image pairs: Setwise active learning for relative attributes. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2856– 2863. [find](https://scholar.google.com/scholar?q=Lucy+Liang+and+Kristen+Grauman%2E+2014%2E+Beyond+comparing+image+pairs%3A+Setwise+active+learning+for+relative+attributes%2E+In+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+pages+2856%E2%80%93+2)
9. L. Plackett, R.1̇975. The analysis of permutations. Journal of the Royal Statistical Society: Series C (Applied Statistics), 24(2):193–202. [find](https://scholar.google.com/scholar?q=L%2E+Plackett%2C+R%2E1%CC%87975%2E+The+analysis+of+permutations%2E+Journal+of+the+Royal+Statistical+Society%3A+Series+C+%28Applied+Statistics%29%2C+24%282%29%3A193%E2%80%93202%2E)
10. Zhen Qin, Rolf Jagerman, Kai Hui, Honglei Zhuang, Junru Wu, Le Yan, Jiaming Shen, Tianqi Liu, Jialu Liu, Donald Metzler, Xuanhui Wang, and Michael Bendersky. 2024. [Large language models are effective text rankers with pairwise ranking prompting](https://md.research.elamur.ai/free/en/papers/qin-2024-large-language-models-are-effective/). In Findings of the Association for Computational Lingu…
11. Lincoln Quillian, Devah Pager, Ole Hexel, and Arnfinn H. Midtbøen. 2017. Meta-analysis of field experiments shows no change in racial discrimination in hiring over time. Proceedings of the National Academy of Sciences, 114(41):10870–10875. [find](https://scholar.google.com/scholar?q=Lincoln+Quillian%2C+Devah+Pager%2C+Ole+Hexel%2C+and+Arnfinn+H%2E+Midtb%C3%B8en%2E+2017%2E+Meta%2Danalysis+of+field+experiments+shows+no+change+in+racial+discrimination+in+hiring+over+time%2E+Proceedings+of+the+National+Ac)
12. Manish Raghavan, Solon Barocas, Jon Kleinberg, and Karen Levy. 2020. Mitigating bias in algorithmic hiring: Evaluating claims and practices. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency (FAT* / FAccT), pages 469–481. ACM. [find](https://scholar.google.com/scholar?q=Manish+Raghavan%2C+Solon+Barocas%2C+Jon+Kleinberg%2C+and+Karen+Levy%2E+2020%2E+Mitigating+bias+in+algorithmic+hiring%3A+Evaluating+claims+and+practices%2E+In+Proceedings+of+the+2020+Conference+on+Fairness%2C+Accounta)
13. H. Sebastian Seung, Manfred Opper, and Haim Sompolinsky. 1992. Query by committee. Proceedings of the Fifth Annual Workshop on Computational Learning Theory (COLT), pages 287–294. [find](https://scholar.google.com/scholar?q=H%2E+Sebastian+Seung%2C+Manfred+Opper%2C+and+Haim+Sompolinsky%2E+1992%2E+Query+by+committee%2E+Proceedings+of+the+Fifth+Annual+Workshop+on+Computational+Learning+Theory+%28COLT%29%2C+pages+287%E2%80%93294%2E)
14. Swanand Vaishampayan, Hunter Leary, Yoseph Berhanu Alebachew, Louis Hickman, Brent Stevenor, Weston Beck, and Chris Brown. 2025. Human and llm-based resume matching: An observational study. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 4808–4823. [find](https://scholar.google.com/scholar?q=Swanand+Vaishampayan%2C+Hunter+Leary%2C+Yoseph+Berhanu+Alebachew%2C+Louis+Hickman%2C+Brent+Stevenor%2C+Weston+Beck%2C+and+Chris+Brown%2E+2025%2E+Human+and+llm%2Dbased+resume+matching%3A+An+observational+study%2E+In+Finding)
15. Yidong Wang, Yunze Song, Tingyuan Zhu, Xuanwang Zhang, Zhuohao Yu, Hao Chen, Chiyu Song, Qiufeng Wang, Cunxiang Wang, Zhen Wu, Xinyu Dai, Yue Zhang, Wei Ye, and Shikun Zhang. 2025. Trustjudge: Inconsistencies of llm-as-a-judge and how to alleviate them. arXiv preprint. ArXiv:2509.21117. — [arXiv:2509.21117](https://arxiv.org/abs/2509.21117) [find](https://scholar.google.com/scholar?q=Yidong+Wang%2C+Yunze+Song%2C+Tingyuan+Zhu%2C+Xuanwang+Zhang%2C+Zhuohao+Yu%2C+Hao+Chen%2C+Chiyu+Song%2C+Qiufeng+Wang%2C+Cunxiang+Wang%2C+Zhen+Wu%2C+Xinyu+Dai%2C+Yue+Zhang%2C+Wei+Ye%2C+and+Shikun+Zhang%2E+2025%2E+Trustjudge%3A+Inconsi)
16. Kamer Ali Yuksel and Hassan Sawaf. 2025. Llm-driven active listwise tournaments for portfolio selection in large asset universes. SSRN Electronic Journal. Preprint; Available at SSRN: [https://ssrn.com/abstract=5764903](https://ssrn.com/abstract=5764903). [↗](https://ssrn.com/abstract=5764903)
17. Yifan Zeng, Ojas Tendolkar, Raymond Baartmans, Qingyun Wu, Lizhong Chen, and Huazheng Wang. 2024. Llm-rankfusion: Mitigating intrinsic inconsistency in llm-based ranking. arXiv preprint (cs.IR). ArXiv:2406.00231. — [arXiv:2406.00231](https://arxiv.org/abs/2406.00231) [find](https://scholar.google.com/scholar?q=Yifan+Zeng%2C+Ojas+Tendolkar%2C+Raymond+Baartmans%2C+Qingyun+Wu%2C+Lizhong+Chen%2C+and+Huazheng+Wang%2E+2024%2E+Llm%2Drankfusion%3A+Mitigating+intrinsic+inconsistency+in+llm%2Dbased+ranking%2E+arXiv+preprint+%28cs%2EIR%29%2E+ArXiv)

*Updated 2 November 2026*