homeprojectsblogcredentialsusesaboutcontact

Projects

tip - click ↻ metrics to flip a card and see live benchmark charts + GitHub stats

01
TPDDL Non-Technical Loss (NTL) Detection Portal

Production-grade FastAPI & AI Command Portal for Tata Power DDL that transforms raw smart meter signals into a probability-ranked enforcement hit-list for targeted electricity theft raids.

PythonFastAPIScikit-LearnMLflowSupabasePlotly.jsPrecision: 73.4%Recall: 97.2%PR-AUC: 0.9696Threshold: 0.6725
2026live ↗gh ↗
metrics · benchmarks
TPDDL Non-Technical Loss (NTL) Detection Portal
Audit Precision
73.4%
Theft Recall Rate
97.2%
PR-AUC Score
96.96%
Audit Precision Floor
65%
73.4%Test Precision> 65% business floor
97.2%Test Recallnear-complete coverage
0.9696PR-AUC Scoreprecision-recall curve
0.6725Optimal Thresholdcalibrated cutoff
02
AutoML: Dataset Cleaner & Analyzer

End-to-end automated ML workflow - dataset upload → cleaning → EDA → training → predictions - with exportable PDF reports and a dataset-aware chatbot.

PythonML PipelineGemini AIHTMLCSSAuto ML PipelinePDF report exportGemini Chatbot
2025live ↗gh ↗
metrics · benchmarks
AutoML: Dataset Cleaner & Analyzer
Pipeline Speedup
85%
Data Quality Score
92%
Bot Response Accuracy
88%
< 5mPipeline Timeaverage dataset
PDFEDA Reportswith comparison charts
SchemaChatbot Contextwith sample rows
ModularPreprocessorsclean, impute, scale
03
YouTube Comment Sentiment Analyzer

Full-stack app that analyzes sentiment of YouTube video comments using a custom-trained LightGBM model exposed via FastAPI, paired with a Chrome Extension for on-the-fly analysis directly from the browser.

PythonLightGBMFastAPINLPChrome ExtensionDockerAccuracy: 79.96%F1: 0.808,869 comments/sec0.11 ms/comment
2025gh ↗
metrics · benchmarks
YouTube Comment Sentiment Analyzer
Overall Accuracy
79.96%
Macro F1-Score
80%
Positive Precision
91%
Negative Recall
80%
8,869Throughputcomments / sec
5.63sPipeline Timefor 50k comments
0.11msAvg Latencyper comment
49,985Test Setlabeled comments
04
Multi-Stock Price Prediction

DL-based stock price forecasting system analyzing 5 years of market data using a 2-layer LSTM with technical indicator feature engineering.

PythonLSTMDeep LearningFastAPIHTMLCSSR² = 0.98Train/test gap: 0.082-layer LSTM
2025live ↗gh ↗
metrics · benchmarks
Multi-Stock Price Prediction
Model R² Score
98%
Train/Test Gap
8%
Feature Importance
75%
5 YrsDataset Spanhistorical market data
2LSTM Layerswith dropout regularization
14msInference TimeFastAPI endpoint speed
MultiTickers Trackeddynamic selection
05
Time Series Demand Forecasting

Industry-grade monthly demand forecasting pipeline across states, districts, and service types with Power BI dashboard integration.

PythonFlaskRandom ForestPower BITime SeriesR² = 0.892MAE: 1842RMSE: 5623
2025live ↗gh ↗
metrics · benchmarks
Time Series Demand Forecasting
Model R² Score
89.2%
Error Reduction
34%
Feature Coverage
95%
1,842Mean Absolute Errordown from 2,790
5,623Root MSEstable variance
ExpandingCV Fold Typeleakage-free walk forward
3-tierGranularitystate, district, service
06
JoSAA College Predictor

ML-based college prediction system analyzing JoSAA counselling data (2021–2025) with XGBoost rank prediction and chance estimation (Safe/Moderate/Risky).

PythonXGBoostMachine LearningData AnalysisHTMLCSS2021–2025 dataXGBoost dual models3-tier chance system
2025live ↗gh ↗
metrics · benchmarks
JoSAA College Predictor
XGBoost Test R²
94.2%
Prediction Precision
89%
Rank Coverage Score
97.5%
5 YrsScraped Range2021 - 2025 counsellings
Dual XGBModel Typeseparate open / close models
3-TierClassificationSafe / Moderate / Risky
100+Institute CountIITs, NITs, IIITs, GFTIs
07
Data Analysis Reports Hub

Centralized multi-page web dashboard showcasing a compilation of comprehensive data analysis reports (India Crime, CPI Inflation, Delhi Rainfall) with interactive Chart.js analytics and role-based controls.

HTML/CSSJavaScriptChart.jsPythonData AnalyticsXGBoostMulti-report compilationXGBoost & ARIMA modelsChart.js Dashboard
2026live ↗gh ↗
metrics · benchmarks
Data Analysis Reports Hub
CPI ARIMA R²
88.4%
Rainfall XGBoost R²
82%
Dashboard Responsiveness
98%
MultiCompilationCrime, CPI, Rainfall
XGBoostRainfall ModelTweedie regression R² 0.82
ARIMAInflation Model156-month time series R² 0.88
VanillaArchitectureHTML/JS + Chart.js Hub
Customer Churn AnalysisSpam Detection (NLP)Sentiment ScannerEDA Automation KitProjects FinderBudget Forecast ToolTask Priority PredictorCarbon Footprint Estimator