Nishat Raihan: Biography, Career, Research and Key Facts

Introduction

Nishat Raihan is a computer science researcher whose work focuses on the rapidly developing fields of artificial intelligence, Natural Language Processing (NLP), Large Language Models (LLMs), and code generation. His research has a particular focus on Code LLMs, multilingual and low-resource programming environments, and the safe use of AI in computer science education. In 2026, he completed his Ph.D. in Computer Science at George Mason University and joined the University of Notre Dame as a Provost’s Postdoctoral Fellow.

His academic profile includes more than 40 publications and more than 700 citations, according to his research profile. His work includes projects such as mHumanEval, TigerLLM, TigerCoder, MojoBench, CSEPrompts, and CodeGuard.

Quick Facts About Nishat Raihan

Who Is Nishat Raihan?

Nishat Raihan is an AI and computer science researcher specializing in language models for code and multilingual natural language processing. His research examines how AI coding systems can be developed for programming languages and human languages that receive less attention in mainstream AI research.

A major theme of his work is accessibility. Instead of focusing only on English and widely used programming languages, he has explored low-resource languages, emerging programming languages, code-mixed text, and other underrepresented settings.

His research also extends into AI safety and education. At Notre Dame, his work involves safety-by-construction guardrails and multilingual program synthesis for Code LLMs.

Early Life and Background

Publicly available professional information about Raihan focuses primarily on his academic and research career rather than his childhood or personal life. He completed his undergraduate education in Bangladesh at the Islamic University of Technology, earning a B.Sc. in Computer Science and Engineering between 2015 and 2018.

His undergraduate record was strong, with a reported GPA of 3.70 out of 4.00 and a seventh-place ranking among 51 students. His early academic work also included research involving optical character recognition and computer vision.

This early interest in computing provided a foundation for his later work in artificial intelligence, natural language processing, and code-generation technologies.

Education and Academic Journey

After completing his bachelor’s degree, Raihan gained professional experience in Bangladesh before moving to the United States for graduate studies.

He joined George Mason University in 2021, where he completed both an M.S. and a Ph.D. in Computer Science. His master’s degree was completed in 2024, while his doctoral degree was completed in 2026.

His reported graduate academic performance was also strong. He earned a 3.97 GPA during his master’s program and a 3.96 GPA during his doctoral studies, along with Distinguished Academic Achievement recognition.

His doctoral dissertation was titled Exploring and Adapting Code LLMs for Underrepresented Domains. The research reflects his continuing interest in adapting AI coding technologies to areas that are not adequately represented by mainstream models.

Career and Major Achievements

Raihan’s career combines software engineering, university teaching, doctoral research, and AI research. Before his Ph.D., he worked at Samsung R&D Institute Bangladesh in software engineering and RPA/OCR development. Later, he became a graduate researcher at George Mason University and, in 2026, joined Notre Dame as a Provost’s Postdoctoral Fellow.

Among his notable achievements are contributions to mHumanEval, TigerLLM, TigerCoder, MojoBench, and CodeGuard. His research profile reports more than 40 publications and more than 700 citations.

Career at Samsung R&D Institute Bangladesh

Before entering his doctoral career in the United States, Raihan worked as a Software Engineer at Samsung R&D Institute Bangladesh in Dhaka from 2019 to 2021.

His professional work involved Robotic Process Automation and Optical Character Recognition. He built RPA bots and services and worked with OCR technologies including Tesseract, Google Vision, ABBYY, Azure, and Tegaki, using Python-based development pipelines.

This industry experience gave him practical exposure to automation and artificial intelligence before he moved deeper into academic research.

Research at George Mason University

At George Mason University, Raihan worked as a Graduate Research Assistant in the Language Technology Lab from January 2023 through July 2026. His research included multilingual NLP, code-mixed text, and Code LLMs for low-resource settings.

He also gained teaching experience as a Graduate Teaching Assistant. His teaching activities included courses related to algorithms, low-level programming, and introductory programming.

His academic work connected research and teaching, including activities in which students evaluated or critiqued AI-generated code.

Current Role at the University of Notre Dame

In August 2026, Raihan joined the University of Notre Dame as a Provost’s Postdoctoral Fellow in the Department of Computer Science and Engineering.

Notre Dame identifies Joanna C. S. Santos as his research advisor. His work is part of the Security & Software Engineering Research Lab, where his research focuses on guardrails and multilingual program synthesis for Code LLMs.

Notre Dame describes his research as bringing AI code generation to programming and human languages that current models often overlook. This includes low-resource languages and emerging programming languages.

Research Interests

The central research interests associated with Raihan include Code LLMs, program synthesis, multilingual NLP, AI safety, computer science education, and benchmarks and datasets.

His work is especially notable for combining these areas rather than treating them as completely separate research problems.

Code LLMs

Code LLMs are artificial intelligence models capable of understanding and generating programming code. Raihan’s research examines how these models can be adapted to programming languages beyond the most commonly supported environments.

His work also investigates evaluation methods, training resources, and benchmarks that can reveal how well models perform across different programming settings.

Multilingual and Low-Resource NLP

Another important area is multilingual NLP, with particular attention to Bangla and code-mixed language.

His publications include research involving Bangla-English-Hindi code-mixed datasets and language models. This work addresses situations in which users naturally combine multiple languages in digital communication.

AI Safety in Education

Raihan also studies the safe use of LLMs in computer science education.

His CodeGuard research examines guardrails for AI coding assistants. According to his research description, the approach reduced harmful code completions by 30% to 65% while preserving legitimate coding assistance.

Notable Research Projects

One of Raihan’s most recognized projects is mHumanEval, a multilingual benchmark designed to evaluate LLMs for code generation.

His research profile describes mHumanEval as containing 836,400 prompts across 204 natural languages and 25 programming languages. The project was published at NAACL 2025.

Another major project is TigerLLM, a family of Bangla large language models presented at ACL 2025. It demonstrates his interest in developing language technologies for Bangla rather than relying exclusively on English-centered resources.

TigerCoder, accepted at LREC 2026, focuses on code generation in Bangla. Meanwhile, MojoBench explores language modeling and benchmarking for the Mojo programming language.

CodeGuard and AI Guardrails

CodeGuard is another important part of Raihan’s research portfolio.

The project, presented in the Findings of EACL 2026, focuses on improving LLM guardrails in computer science education. It combines a prompt taxonomy, an 8,000-prompt dataset, and PromptShield to identify and reduce harmful code completions.

This research is particularly relevant as AI coding assistants become more common in classrooms. The challenge is to prevent unsafe outputs without unnecessarily blocking legitimate programming help.

Publications and Research Impact

Raihan’s publication record covers NLP, Code LLMs, AI safety, computer science education, computer vision, and related areas.

His publication page currently lists more than 40 publications, including journal articles, conference papers, workshop papers, and preprints. It also reports more than 700 citations, an h-index of 14, and an i10-index of 19.

His research has appeared in venues including ACL, NAACL, EACL, EMNLP, LREC, SIGCSE, IEEE BigData, and other conferences and journals.

Awards and Academic Recognition

Raihan has received several academic distinctions during his graduate career.

George Mason University awarded him a Doctoral Research Scholarship, Tier 1, in 2026, and he also received a Distinguished Academic Achievement Award for his Ph.D. His master’s studies brought another Distinguished Academic Achievement Award in 2024.

His research also received a Best Paper Award at the first Workshop on Bangla Language Processing held in connection with EMNLP 2023.

Age and Physical Features

A reliable public source does not provide a confirmed date of birth or detailed physical measurements for Raihan. Therefore, his exact age, height, weight, and other physical characteristics should not be stated as facts without reliable documentation.

His professional profiles are primarily devoted to his education, research interests, publications, and academic appointments rather than personal physical details.

Relationship With and Marriage and Family Life

Raihan keeps his professional profile focused on academic and research activities. Available public sources do not provide verified information about a spouse, marriage, romantic relationship, or detailed family life.

For this reason, claims about his relationship status or private family circumstances should be treated cautiously unless they come directly from a reliable public source.

Personality and Interests

Raihan’s public academic work suggests a strong interest in developing practical and accessible AI technologies.

His research repeatedly returns to underrepresented languages, programming environments, educational applications, and openly available benchmarks. He has also contributed to academic teaching, mentoring, research organization, and professional reviewing.

Rather than focusing exclusively on model development, his work often emphasizes evaluation and resources that allow other researchers to measure AI systems more effectively.

Net Worth and Income

There is no reliable public information establishing Raihan’s personal net worth, investments, or total income.

As an academic researcher, his professional earnings are associated with university appointments and research positions, but a precise salary or net worth figure has not been publicly established in the sources reviewed.

Therefore, online estimates of his net worth should not be presented as confirmed facts.

Social Media Presence

Raihan maintains a public online research presence through his professional website and academic research platforms.

His public profile includes information about his publications, research projects, datasets, academic appointments, talks, and current research directions. His work is also represented through research and dataset repositories used by the academic and AI communities.

His online presence is primarily professional rather than centered on entertainment or celebrity-style social media activity.

Public Image and Legacy

Raihan’s emerging academic profile is closely associated with research into underrepresented areas of AI and programming.

His work on Bangla language models, multilingual code generation, Code LLM benchmarks, and AI guardrails reflects a broader effort to make AI systems more useful across languages and technical environments that receive less attention.

Projects such as mHumanEval and TigerCoder also emphasize the importance of evaluation. By creating benchmarks and datasets, researchers can better understand where current AI systems perform well and where they still have limitations.

Because his academic career is still developing, his long-term legacy cannot yet be determined. However, his existing research provides a substantial foundation for continued work in Code LLMs, multilingual AI, and AI safety.

Conclusion

Nishat Raihan has built an academic career around some of the most important questions in modern artificial intelligence. From his early work in software engineering and OCR at Samsung R&D Institute Bangladesh to his doctoral research at George Mason University, his career has gradually moved toward language models, code generation, multilingual NLP, and AI safety.

In 2026, he completed his Ph.D. and joined the University of Notre Dame as a Provost’s Postdoctoral Fellow. His current research continues to explore safer and more inclusive Code LLMs, particularly for languages and programming environments that are often overlooked.

With more than 40 publications and more than 700 reported citations, his research portfolio already spans a wide range of AI topics. His continuing work on benchmarks, datasets, multilingual models, and educational guardrails makes him a researcher to watch as the field of AI-assisted programming continues to evolve.

8 FAQs About Nishat Raihan

1. Who is Nishat Raihan?

Nishat Raihan is a computer science researcher specializing in Code LLMs, NLP, multilingual AI, program synthesis, and AI safety.

2. Where does Nishat Raihan work?

He is a Provost’s Postdoctoral Fellow in the Department of Computer Science and Engineering at the University of Notre Dame.

3. Where did Nishat Raihan complete his Ph.D.?

He completed his Ph.D. in Computer Science at George Mason University in 2026.

4. What is Nishat Raihan’s research focus?

His research focuses on Code LLMs, program synthesis, multilingual and low-resource NLP, AI safety, computer science education, and benchmarks and datasets.

5. What is mHumanEval?

mHumanEval is a multilingual benchmark designed to evaluate large language models for code generation across numerous natural and programming languages.

6. What is TigerLLM?

TigerLLM is a family of Bangla large language models developed as part of Raihan’s research on multilingual and low-resource language technologies. It was presented at ACL 2025.

7. Did Nishat Raihan work in industry?

Yes. He worked as a Software Engineer at Samsung R&D Institute Bangladesh from 2019 to 2021, focusing on RPA and OCR development.

8. How many publications does Nishat Raihan have?

His current research profile lists more than 40 publications and reports more than 700 citations. These figures can change as new research is published and citations accumulate.

People Bio