# Manan Dudeja

Manan Dudeja is a remote worker on Himalayas, Research Intern at Samsung PRISM, based in India.

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

```text
AI researcher at Samsung PRISM working on low-light camera enhancement, with prior multimodal weed segmentation research achieving 91.33% accuracy.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | India |
| Authorized to work in | India |
| Salary expectations | 1027k+ INR |
| Job categories | Machine Learning Engineer, AI Researcher, Research Scientist |
| Skills | Machine Learning, Deep Learning, Computer Vision, Multimodal Learning, Image Quality Enhancement, Image Segmentation, Natural Language Processing, Large Language Models, RAG, Agentic AI, Data Analysis, Statistical Analysis, Feature Engineering, Model Evaluation, Experimentation |
| Portfolio | https://github.com/dudejamanan |
| LinkedIn | https://linkedin.com/in/dudejamanan |
| GitHub | https://github.com/dudejamanan |

## About

```text
At Samsung PRISM, I’m researching low-light image enhancement and single-frame fusion for smartphone camera previews, with a focus on inference speed and compute constraints. I’m reproducing LiteIE and benchmarking exposure, denoising, detail preservation, and efficiency.

At IIT Ropar, I developed multimodal weed segmentation approaches, including a lightweight UAV model that achieved 91.33% accuracy. I also co-founded Audaxity, where I built speech-recognition and LLM pipelines for structured, syllabus-aligned lecture notes.
```

## Experience

### Research Intern (current)

Samsung PRISM. Sep 2026 - Present

```text
Researching low-light image enhancement and single-frame fusion for smartphone camera previews, targeting ≥24 FPS under tight memory and compute constraints. Reproducing LiteIE and benchmarking exposure, denoising, detail preservation and inference efficiency using PSNR, SSIM and NIQE.
```

Skills: Low-Light-Enhancement, Image-Fusion, PSNR, SSIM, NIQE, Benchmarking, Computer-Vision

### Head of Research & Development (current)

HackClub, VIT Chennai. May 2026 - Present

```text
Lead AI/ML initiatives and mentor project teams; organized workshops and events with 400+ participants.
```

Skills: AI, Mentorship, Event-Organization, Leadership

### Co-Founder & CTO (current)

Audaxity. Jan 2026 - Present

```text
Built ASR + LLM pipelines turning lecture audio into structured, syllabus-aligned notes; benchmarked 4 ASR systems (Whisper large-v3, Deepgram Nova-3, AssemblyAI, Sarvam Saaras) to select the best model, and designed RAG Q&A with pgvector.
```

Skills: ASR, LLM, RAG, Pgvector, Benchmarking, Python, fastAPI

### Research Intern

IIT Ropar. May 2026 - Jul 2026

```text
Formulated an RGB + multispectral segmentation approach using vegetation indices, multi-level feature fusion and CBAM attention; achieved 80.79% IoU and 89.21% Dice. Benchmarked against 11 CNN, Transformer and hybrid models; ran ablation studies and hyperparameter optimization to validate each design choice; manuscript submitted to Artificial Intelligence Review. Built a lightweight dual-branch mo
```

Skills: Deep-Learning, Computer-Vision, Image-Segmentation, Multispectral-Imaging, CBAM, Ablation-Studies, Hyperparameter-Optimization, PyTorch, TensorFlow

### Gen AI Associate

Dezors. Jan 2026 - Apr 2026

```text
Contributed to an AI document-generation platform: implemented partial editing and versioning, and supported production deployment with Docker and Kubernetes.
```

Skills: AI, Document-Generation, Docker, Kubernetes, Python

## Education

### Vellore Institute of Technology

B.Tech, Computer Science and Engineering. 2024 - Present

```text
Pursuing a B.Tech in Computer Science and Engineering with coursework in data structures, operating systems, machine learning, and deep learning.
```

## Tech stack

- Kubernetes
- NumPy
- Pandas
- PostgreSQL
- OpenCV
- JavaScript
- Python
- Java
- C++
- C
- TensorFlow
- PyTorch
- scikit-learn
- FastAPI
- Transformers
- Git
- Docker
- SQL
- LangChain
- pgvector
- Faiss
- LangGraph
- Seaborn
- Matplotlib

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

## People also viewed

- [Richard Abishai](https://himalayas.app/@richardabishai): Research Student, University of Malaya
- [Kumarran Mahesh](https://himalayas.app/@kumarran): Club Lead, IOTA (Projects Club)
- [Tanish User](https://himalayas.app/@tanishuser1): Developer Intern, Samsung R&D Institute India
- [Debanjan Chowdhury](https://himalayas.app/@debanjanchowdhury): Senior Deep Learning Engineer, Destar Inc.
- [shivali bhatnagar](https://himalayas.app/@shivalibhatnagar): Senior Engineer, Samsung Research Institute Bangalore
- [Prateek Singh](https://himalayas.app/@prateeksingh7): Research Intern, Indian Institute of Technology Ropar
- [Siddharth Upl](https://himalayas.app/@siddharthupl): Chair, IEEE Computer Society Student Chapter, Mahindra University
- [MANSI GOEL](https://himalayas.app/@mansigoel): Senior Software Engineer – AI/ML & Agentic Systems, NVIDIA
- [MD Gulam Hussain](https://himalayas.app/@mdgulamhussain): Academic Research Trainee, National Institute of Technology (NIT), Warangal
- [Nakul Sinha](https://himalayas.app/@nakulsinha): Student Researcher, Samsung SEED Lab
