Leibniz University, Germany
Jan 2020 – Jun 2020Developed ML models to predict dementia risk with sparse, high-dimensional data and designed an automata-based graph simulation model for COVID-19 spread analysis and forecasting.
Senior Data Scientist · Bangalore, India
William Scott Paka
Machine learning at scale, at Walmart
Senior Data Scientist with 8+ years of experience in natural language processing, machine learning, deep learning, recommendation engines and computer vision. M.Tech in Artificial Intelligence from IIIT Delhi. Published 5 papers and holds 5 patents building state-of-the-art models in NLP.
- Years of Experience
- 8+
- Publications
- 5
- Patents
- 5
01
Experience
-
Walmart
Oct 2024 – PresentSenior Data Scientist
-
PromptCloud Technologies
Dec 2021 – Oct 2024Senior Data Scientist
-
KaHa Technologies (Cove IoT)
Aug 2021 – Dec 2021Senior Data Scientist
-
TCS Research (TRDDC)
Sep 2020 – Aug 2021Deep Learning & AI Researcher
-
Covenant IT Solutions
Mar 2016 – Jun 2018Data Scientist
Internships — Data Science
UnitedHealth Group (Optum Global Services)
Apr 2019 – Aug 2019Developed and deployed an end-to-end LDA-based model to automate detection of fraudulent medical discharge summaries, presented at LIC Delhi and currently in use by the NHA. Enhanced LDA for short texts by 5–8% through sentiment analysis.
02
Publications & Patents
-
ENDEMIC: a semi-supervised co-attention network for early detection of COVID-19 fake news
Achieved 97% accuracy in early detection of COVID-19 fake news.
PAKDD 2021 -
Cross-SEAN: a cross-stitch semi-supervised neural attention model for COVID-19 fake news detection
Achieved 95% accuracy in COVID-19 fake news detection.
Elsevier 2021 -
Embedding-based joint-sentiment model for short text (Master's thesis)
Average improvement of 10% in topic coherence and 5% in topic diversification.
ICWSM 2021 -
Blackmarket tweet detection
Model classifying the source of tweets (blackmarket vs genuine), F1-score 0.89.
ASONAM 2019 -
Conference task on multi-modal data
Conducted a conference task on multi-modal data (textual, visual & audio) with 50+ participants.
SemEval-2020
Patents — 5 total
3 filed & 1 approved at Walmart, in collaboration with the team.
03
Skills
Programming
- Python
- Java
- C
- C++
ML / AI
- NLP
- Machine Learning
- Deep Learning
- GenAI
- LLMs
- Computer Vision
- Recommendation Engines
- MLOps
- Time Series
- Statistics
- Data Analysis
Frameworks & Tools
- PyTorch
- TensorFlow
- Keras
- PySpark
- Scikit-learn
- NLTK
- SpaCy
- OpenCV
- Seaborn / Matplotlib
- Flask
- MongoDB
- SQL
- Spark / Hadoop
- AWS
- Docker
- Git
- Jupyter
Core CS
- OOPs
- Data Structures & Algorithms
- Problem Solving
04
Education & Certifications
Education
IIIT Delhi
M.Tech, Artificial Intelligence · Aug 2020
Karunya University, Coimbatore
B.Tech, Computer Science and Engineering · Aug 2016
Certifications
- Machine LearningStanford University
- Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and OptimizationDeepLearning.AI
- Neural Networks and Deep LearningDeepLearning.AI
- Structuring Machine Learning ProjectsDeepLearning.AI
05
Achievements
Teaching Assistant, Machine Learning (PG), IIIT Delhi, 2019.
Ranked 14 in the hard-level HackerRank Machine Learning Challenge "Guess the Flipkart Query".
2nd prize, "King Coders" (2016) — coding event by the Computer Society of India.
2nd prize, COD++ (2015) — coding event at national-level technical symposium "Expertise 2k15".
Let's talk.
Open to conversations about machine learning, NLP research, and building recommendation systems at scale.