# Sankalita Paul

Sankalita Paul is a remote worker on Himalayas, Junior Research Associate-Data Strategy and AI at Mestastop Solutions Pvt. Ltd., based in India.

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

```text
Data and ML researcher at Mestastop Solutions developing oncology pipelines that predict metastasis from whole-slide images and clinical data.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | India |
| Authorized to work in | India |
| Salary expectations | 845k+ INR |
| Job categories | Machine Learning Engineer, Data Scientist, Bioinformatics Scientist, Research Scientist |
| Skills | Machine Learning, Deep Learning, Multiple Instance Learning, Whole Slide Image Processing, Medical Imaging, Clinical Data Analysis, Oncology, Metastasis Prediction, Multimodal Fusion, Attention Mechanisms, Convolutional Neural Networks, Transformer Models, Model Explainability, Genomics, Transcriptomics |

## What I'm looking for

```text
I'm looking to build on my experience developing end-to-end data and machine learning pipelines for clinical and biological datasets, especially for oncology applications.
```

## About

```text
At Mestastop Solutions Pvt. Ltd., I build machine learning pipelines for oncology applications using patient-level clinical data and whole-slide images. I developed a preprocessing pipeline for histopathological images, covering tissue detection, patch extraction, scanner variability, and slide-level quality control.

I developed a Multiple Instance Learning framework with attention-based aggregation to predict metastasis from weakly labelled whole-slide images. It uses convolutional and transformer-based encoders to extract patch-level features.

I’m extending the pipeline to combine structured clinical data with whole-slide image embeddings through late fusion. I also use attention heatmaps to identify tissue regions correlated with metastatic outcomes.

In my Oncology Risk Intelligence portfolio project, I developed and deployed a Streamlit app for cancer metastasis risk prediction using Random Forest models and SHAP explainability. My academic projects included modeling comorbidity risks in autistic individuals and classifying brain tumours from MRI images.
```

## Experience

### Junior Research Associate-Data Strategy and AI (current)

Mestastop Solutions Pvt. Ltd.. Aug 2025 - Present

```text
Developed end-to-end WSI preprocessing pipelines and a Multiple Instance Learning framework for metastasis prediction from whole-slide images. Extended the pipeline to multi-modal fusion with clinical data and interpreted model outputs using attention heatmaps.
```

Skills: Python, PyTorch, TensorFlow, OpenCV, OpenSlide, HDF5, Multiple-Instance-Learning, Attention-Mechanisms, Convolutional-Neural-Networks, Transformers, data-preprocessing, Quality-Control, Multimodal-Fusion, SHAP

## Education

### Institute of Bioinformatics and Applied Biotechnology

Master of Science, Big Data Biology. 2023 - 2025

```text
Masters in Big Data Biology from the Institute of Bioinformatics and Applied Biotechnology, Bangalore, from July 2023 to May 2025.
```

### Calcutta University

Bachelor of Science (Honours), Microbiology. 2020 - 2023

```text
B.Sc (Hons) in Microbiology from Calcutta University, from September 2020 to June 2023.
```

### Coursera

Advanced Learning Algorithms, Machine Learning

```text
Completed the Advanced Learning Algorithms certification on Coursera.
```

### Coursera

Unsupervised Learning Algorithms: Recommenders, Reinforcement Learning, Machine Learning

```text
Completed the Unsupervised Learning Algorithms: Recommenders, Reinforcement Learning certification on Coursera.
```

## Tech stack

- GitHub
- NumPy
- Pandas
- OpenCV
- Python
- TensorFlow
- PyTorch
- scikit-learn
- Streamlit
- Linux
- Git
- SQL
- Shiny
- Seaborn
- Matplotlib

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

## People also viewed

- [John F Aradan](https://himalayas.app/@johnfaradan): PennApps Hackathon Participant Hacker, University of Pennsylvania
- [sriram User](https://himalayas.app/@sriramuser1): Research Intern, Indian Institute of Science
- [Krishna preethi Appana](https://himalayas.app/@krishnapreethiappana): Data Science Intern, CODSOFT
- [Akshat Pandey](https://himalayas.app/@akshatpandey1): Research Associate, IIT Jodhpur
- [Priyam Shree](https://himalayas.app/@priyamshree): Research Intern, National Institute of Technology, Rourkela
- [Shubham Mahindrakar](https://himalayas.app/@shubhammahindrakar): AI Bioinformatics Research Intern, EFREI Lab
- [Md Ashraf](https://himalayas.app/@mdashraf3): ML Intern, Ministry of Electronics and Information Technology (MeitY) &
- [Khushi Singh](https://himalayas.app/@khushisingh9): Data Analyst Intern, Unified Mentor
- [Girija Awate](https://himalayas.app/@girijaawate): Research Intern Deep Learning & Medical Imaging, Indian Institute of Technology (IIT) Bombay
- [Chanchal Gupta](https://himalayas.app/@chanchalgupta3): Research Assistant, Medal Lab, IIT Bombay
