Shikhar Chauhan
Data Scientist
Antwerp
Summary
Experienced data scientist with 5+ years of expertise in end-to-end development of machine learning algorithms, encompassing scoping business needs, data collection, analysis, modeling, and large-scale deployment. Specialized in NLP, Computer Vision, and multi-modal deep learning. Passionate about developing performant and reusable solutions. Committed to continuous learning and sharing knowledge.
Overview
7
7
years of professional experience
7
7
years of post-secondary education
3
3
Languages
Work History
Data Scientist
KBC Bank N.V
Leuven
12.2021 - Current
- Collaborated with cross-functional teams to design and implement robust, end-to-end document processing pipelines for insurance invoice payment automation.
- Spearheaded the development of advanced multi-modal document representation and deep learning libraries, resulting in a remarkable >21% increase in STP rates, >40% faster inference times.
- Worked with stakeholders to develop quarterly roadmaps based on impact, effort and test coordinations, and supervise team of 4 data scientists that drove significant business value.
Research Intern
02.2021 - 04.2021
- Research Internship as part of my thesis
- Researched into the importance of multi modal information(textual + visual + spatial) for Key
- Information Extraction from documents
- Worked on representing documents as graphs, and enriching embeddings with multi modal information, that help improve performance in downstream tasks.
Researcher
Imec
Antwerp
09.2019 - 08.2020
- Researching deep learning models that help with automation of recycling electronic waste
- My prime focus of research is making deep learning effective with Class Imbalance, and model distillation and compression
- Implemented Active Learning algorithms for effective data labelling.
- Experimented widely with model explainability methods that allow us to focus on the gaps in learning representations
- Co-founded IDLab Nexus, and inter-lab meetup for researchers at IDLab where they share their research and interests.
Machine Learning Engineer
Factmata
09.2018 - 07.2019
- Undertook initiative to develop data exploration modules that helps in drawing insights from datasets which led to better understanding towards gaps and biases in datasets
- Used model interpretability techniques to explore the biases and features models learn
- Developed classifiers to detect toxic content online
- Worked on building a trend detection system using unsupervised learning
- Data Pipelines that scrape URLs and score them using machine learning models
- Developed utilities and toolkits for data preprocessing, training models, and collecting data.
Machine Learning Engineer
98Fit
New Delhi
01.2017 - 08.2018
- Developing Machine Learning algorithms to develop Diet plans and Workout plans for users based on their lifestyles and bodies using PyTorch and Genetic Algorithms.
- End to End integration of Machine Learning Models with Backend.
Contributor
Facebook Open Source Mentorship
04.2018 - 06.2018
- Facebook Open Source Mentorship Program is a 6 week program where I worked on open source projects under a Facebook Engineer
- Visdom is a flexible tool for creating, organizing, and sharing visualizations of live, rich data
- I worked on adding a read-only mode to the visdom server
- I worked on adding CPU based gradient for ROIAlign to pytorch/vision.
Machine Learning and Backend Engineer
Neuron Inc
11.2016 - 01.2017
- Developed a prototypal Chatbot Framework built upon Recurrent Neural Networks and Non
- Deterministic Finite Automatons
- Developed APIs for interaction with Chatbot
Full Stack Developer
Edunuts
New Delhi
12.2015 - 09.2016
Education
Advanced Master - Artificial Intelligence
Katholieke Universiteit Leuven
Leuven 09.2020 - 08.2023
Bachelor of Technology - Mathematics And Computing Engineering
Delhi Technological University
New Delhi, India 08.2013 - 07.2017
Deep Learning Specialisation -
Deeplearning.ai
Skills
PyTorch
TensorFlow
Docker
Scikit
Pandas
Numpy
Python
SQL
Git
Spacy
HuggingFace
PySpark
Linux
AWS
TorchScript/TorchServe
Flask / Fast API / Django
Accomplishments
- World Finalist Facebook Hackathon 2016.
- Contributor to pytorch/vision, huggingface/transformers, spaCy, pytorch-lightning.
Timeline
Data Scientist
KBC Bank N.V
12.2021 - Current
Research Intern
02.2021 - 04.2021
Advanced Master - Artificial Intelligence
Katholieke Universiteit Leuven
09.2020 - 08.2023
Researcher
Imec
09.2019 - 08.2020
Machine Learning Engineer
Factmata
09.2018 - 07.2019
Contributor
Facebook Open Source Mentorship
04.2018 - 06.2018
Machine Learning Engineer
98Fit
01.2017 - 08.2018
Machine Learning and Backend Engineer
Neuron Inc
11.2016 - 01.2017
Full Stack Developer
Edunuts
12.2015 - 09.2016
Bachelor of Technology - Mathematics And Computing Engineering
Delhi Technological University
08.2013 - 07.2017
Deep Learning Specialisation -
Deeplearning.ai