
Learning new everyday!
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
MNIST-Digit-Classifier
June 4, 2026 – Present
A PyTorch-based MNIST handwritten digit classifier built from first principles to understand deep learning fundamentals, neural networks, and multiclass image classification.
View ProjectDataPrep-Workflow-Engine
June 2, 2026 – Present
An interactive Streamlit-based workflow engine for Exploratory Data Analysis (EDA), sequential data preprocessing, and cleaned dataset export for Machine Learning pipelines.
View ProjectFMCG-Purchase-Propensity-Prediction
May 26, 2026 – Present
FMCG Purchase Propensity Prediction Dashboard — A machine learning-powered campaign simulation system that predicts customer purchase likelihood and expected demand using customer behavior analytics, XGBoost classification, and business-driven inference workflows.
View ProjectFMCG-Sales-Potential-Prediction
May 25, 2026 – Present
A Machine Learning-powered Streamlit dashboard that predicts the potential sales performance of FMCG products based on product characteristics, outlet features, and historical retail sales patterns using regression models like XGBoost, Random Forest, and Linear Regression.
View ProjectLean-Body-Mass-Prediction
May 19, 2026 – Present
Machine Learning project for Lean Body Mass prediction using feature engineering, XGBoost, cross-validation, and hyperparameter tuning.
View ProjectGHG-Emissions-Model-Training
May 16, 2026 – Present
Machine Learning project for predicting greenhouse gas emissions using EDA, regression models, and model evaluation techniques.
View ProjectPortfolio-Pawan
May 16, 2026 – Present
Personal portfolio built with a minimal engineering-focused aesthetic to showcase projects, systems thinking, and software development experience.
View ProjectDevWatch
March 2, 2026 – Present
DevWatch – Deep Developer-Aware System Dashboard for Linux. A GNOME extension that gives “project-aware system intelligence” — it shows processes, ports, performance, conflicts, resource usage, and session restore directly in the desktop panel.
View ProjectCodeCrypt
February 3, 2026 – Present
CodeCrypt – Secure Anti-Cheat Assessment Platform. A React + TypeScript web app that enforces fullscreen exams with tab-switch detection, real-time violation tracking, warning system, and automatic lockout for secure online testing.
View ProjectFinTwitch
October 4, 2025 – Present
FinTwitch – Gamified Personal Finance Platform. A full-stack web app built with React and Firebase that combines finance tools, habit tracking, quizzes, and stock insights into an interactive, game-like experience for smarter money management.
View ProjectCultural Fit Analysis
The candidate's portfolio demonstrates a strong inclination towards personal projects, indicating self-motivation and initiative. The projects span web development, system utilities, and machine learning, suggesting a broad interest in technology. For a 'Data Scientist' role, the machine learning projects are highly relevant, but the significant number of non-data science projects (e.g., CodeCrypt, DevWatch, FinTwitch) might indicate a broader interest beyond a focused data science career path, or a strong full-stack/software engineering inclination. This breadth could be a strength for cross-functional teams but might also suggest less specialized depth in core data science areas compared to a candidate with a more focused portfolio. The lack of team-based projects or professional experience makes it difficult to assess collaboration fit.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a proactive and self-directed approach to learning and problem-solving. The diversity of projects suggests adaptability and a willingness to explore different technical domains. However, without specific psychometric or English test scores, a detailed assessment of communication, logical reasoning, work attitude, stress handling, and team collaboration is not possible.