Adwait Tiwari

Data Scientist

Pune, IN.

About

Highly analytical Data Scientist with 1+ years of experience in developing and deploying advanced Machine Learning and AI models, specializing in fraud detection, anomaly detection, and recommendation systems. Proven expertise in Python, deep learning frameworks, statistical modeling, A/B testing, and end-to-end MLOps. Adept at leveraging statistical analysis, feature engineering, and model optimization to translate complex ML solutions into tangible business value and drive data-driven strategies.

Work

Cheqlt

|

Data Scientist

→

Summary

As a Data Scientist at Cheqlt, I focus on developing and deploying advanced machine learning solutions to combat counterfeiting and enhance operational efficiency across supply chain and logistics domains.

Highlights

Engineered an XGBoost-based counterfeit detection system leveraging supply chain and geospatial data, increasing fraud identification by 35%.

Developed a real-time anomaly detection engine with LSTM autoencoders for logistics data, enhancing supply chain visibility by 40%.

Created a CNN-based image recognition system for product verification from user photos, boosting automated verification accuracy by 15%.

Applied DBSCAN clustering to identify geographic counterfeit hotspots from scan data, improving enforcement action success by 25%.

Performed large-scale feature engineering and hyperparameter tuning, optimizing model performance and reducing prediction latency by 20%.

Implemented SHAP and LIME for model interpretability, effectively communicating complex predictions to non-technical stakeholders and informing data-driven strategies.

Education

IIIT Bangalore

→
Executive Diploma Machine Learning & Artificial Intelligence

BMS College of Engineering

→
B.E. Civil Engineering Grade: 8.02/10.0

Skills

Programming Languages & Frameworks

Python, FastAPI, Flask.

AI & Machine Learning

Machine Learning (ML), Deep Learning (DL), Natural Language Processing (NLP), PyTorch, Transformers, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), XGBoost, Random Forest, Decision Tree, LSTM Autoencoders, Convolutional Neural Networks (CNN).

NLP Libraries

Spacy, NLTK.

Data Analysis & Databases

MySQL, MongoDB, Advanced SQL, Pandas, NumPy, Tableau.

Model Deployment & MLOps

MLOps, Streamlit, Docker, Git/GitHub, API Development.

Cloud Platforms & Collaboration

Google Cloud Platform (GCP), Amazon Web Services (AWS), Google Colab.

Statistical Analysis & Feature Engineering

Statistical Modeling, A/B Testing, Feature Engineering, Hyperparameter Tuning, DBSCAN Clustering, ANOVA/Chi2, VIF, KNNImputer.

Model Interpretability

SHAP, LIME.

Projects

Credit Risk Modelling

→

Summary

Developed a robust credit risk pipeline using Python and Scikit-learn, implementing KNNImputer for data imputation and feature selection preparation.

Amazon Sales Analysis (Advanced SQL)

→

Summary

Designed and executed advanced SQL-based sales trend analysis to identify high-performing products and underperforming categories.

Retail App Analytics

→

Summary

Automated a retail data pipeline processing over 10,000 records (CSV/MongoDB) using Pandas and NumPy, storing results in MongoDB for analytics.

Movie Recommender System

→

Summary

Developed a full-stack movie recommender system with Python/Flask backend and JavaScript frontend, integrating the TMDB API for comprehensive movie data.