# Ubaid Ul Hasan

Ubaid Ul Hasan is a remote worker on Himalayas, University Ambassador at Haya Foundation, based in Pakistan.

Source: https://himalayas.app/@ubaidulhasan

```text
AI researcher studying LLM truth probes, whose submitted work investigates entity-frequency confounding in hidden-state evaluations.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | Pakistan |
| Authorized to work in | Pakistan |
| Salary expectations | 30k+ USD |
| Job categories | AI Researcher, Machine Learning Engineer, Computer Vision Engineer |
| Skills | Model Interpretability, LLM Evaluation, AI Safety, Explainable AI, Statistical Analysis, Anomaly Detection, Machine Learning, Deep Learning, Natural Language Processing, Computer Vision, Large Language Models, Python, C++, SQL, PyTorch |
| LinkedIn | https://linkedin.com/in/ubaid-ul-hasan-5b8b6034b |
| GitHub | https://github.com/Ubaid-Ul-Hasan |

## What I'm looking for

```text
I aim to contribute original research that makes increasingly capable AI systems more interpretable, reliable, and safe.
```

## About

```text
My submitted research, “Familiar Means True: Entity-Frequency Confounding in Hidden-State Truth Probes for Large Language Models,” examines whether hidden-state truth probes rely on entity-frequency signals rather than truth-related representations. I contributed to its methodology, experimental analysis, and code development.

In that study, I evaluated 18 language models on 608 statements, examining probe performance when truth labels and entity frequency aligned or conflicted. I also investigated a training-free approach using iterative nullspace projection and pretraining-corpus frequency information to reduce entity-frequency confounding.

I developed a machine-learning system for zero-day attack detection using anomaly-detection methods and explainable analysis. My projects also include a YOLOv8 hard-hat detector, an explainable supply-chain intelligence platform, a local multimodal intent pipeline, and an offline agricultural knowledge assistant.
```

## Experience

### University Ambassador (current)

Haya Foundation. Jan 2026 - Present

```text
Represent the foundation within the university community and engage fellow students. Educate students about applications of artificial intelligence in environmental protection and sustainability.
```

Skills: Communication-Skills, Leadership, AI-For-Sustainability

### Lead Researcher / Developer

Machine Learning-Based Zero-Day Attack Detection System

```text
Developed a machine-learning system to identify potentially unseen cyberattacks through anomaly-detection methods and explainable analysis.
```

Skills: Machine-Learning, Anomaly-Detection, SHAP, Autoencoder, Isolation-Forest, CTGAN, Python

### AI Researcher

Emerson University

```text
Conducted research on mechanistic interpretability and AI safety, focusing on entity-frequency confounding in hidden-state truth probes for large language models.
```

Skills: Model-Interpretability, LLM-Evaluation, Python, PyTorch, Statistical-Analysis

## Education

### Aspire Leaders Program

Aspire Leaders Program, Leadership. 2026 - Present

```text
Participant in Cohort 4 of the Aspire Leaders Program, ongoing since 2026.
```

### Emerson University

Bachelor of Science, Artificial Intelligence. 2024 - Present. Grade: 3.29/4.00

Activities and societies: CM Punjab Honhaar Scholarship

```text
Pursuing a BS in Artificial Intelligence, currently in the 5th semester with a CGPA of 3.29/4.00. Awarded the CM Punjab Honhaar Scholarship.
```

### Elsevier

Certified Peer Reviewer Course, Peer Review

```text
Completed the Elsevier Certified Peer Reviewer Course.
```

### Microsoft

AI Skills Passport, Artificial Intelligence

```text
Completed the Microsoft AI Skills Passport.
```

### MathWorks

MATLAB Onramp, MATLAB

```text
Completed the MATLAB Onramp course.
```

## Tech stack

- GitHub
- NumPy
- Pandas
- OpenCV
- Python
- C++
- MATLAB
- TensorFlow
- PyTorch
- scikit-learn
- Streamlit
- FastAPI
- Git
- SQL
- Google Colab
- MediaPipe

To contact Ubaid, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Sheraz Ali](https://himalayas.app/@sherazali2): System Specialist ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING, State Life Insurance Corporation (SLIC)
- [Hasnain Ahmad](https://himalayas.app/@hasnainahmad): Deep Learning Researcher, Abdul Wali Khan University Mardan
- [Farah Saeed](https://himalayas.app/@farahsaeed): Doctoral Scholar, University of Georgia
- [Shaheer Abbas](https://himalayas.app/@shaheerabbas): Event Coordinator, IEEE GIK Chapter
- [Emaan Arshad](https://himalayas.app/@emaanarshad): Summer Intern, State Bank of Pakistan
- [Syble Jan](https://himalayas.app/@syblejan): AI/ML Intern, Developer Hubs
- [Muhammad Abdullah](https://himalayas.app/@muhammadabdullah68): AI Team Lead, Machine Vision & Intelligent Systems Lab
- [Abdul Rafay Bajwa](https://himalayas.app/@abdulrafaybajwa): Day to Day Supervisor, LUMS
- [Areej Waseem](https://himalayas.app/@areejwaseem): Image Detector Model Labelling, Unknown
- [Ayesha Irshad](https://himalayas.app/@ayeshairshad): Research Team Lead, Codet.ai
