Skip to content
View Saksham3124's full-sized avatar
1 follower · 5 following

Highlights

  • Pro

Block or report Saksham3124

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Saksham3124/README.md

Hi 👋, I'm Kumar Saksham

Data Analyst | B.Tech in ECE, BIT Mesra (2026)

Turning raw data into insights through analytics, SQL, and dashboards.


🚀 About Me

  • 🔭 Built end-to-end analytics systems in compliance, operations, and finance — actively seeking analyst roles
  • 🤝 Looking to collaborate on data analysis, business intelligence, SQL, or dashboard projects
  • 📍 Open to Data Analyst roles
  • 💬 Ask me about SQL window functions, anomaly detection, Power BI, Tableau, Python for data analysis
  • ⚡ Fun fact: I built a 3-layer GST fraud detection system that flags 15% of 50,000 invoices — using pure SQL and statistics, no ML

🛠️ Tech Stack

Languages

Python SQL

Data & Analytics Libraries

Pandas Scikit-learn NumPy

Databases

PostgreSQL SQLite

BI & Visualization

Tableau Power BI

Tools

Git GitHub Excel

📊 Featured Projects

🔍 GST Invoice Anomaly Detection & Vendor Risk Scoring

End-to-end analytics system detecting fraudulent GST invoice patterns across 50,000+ records

  • 3-layer detection pipeline — Rule-based validation → Statistical anomaly detection → Weighted risk scoring
  • SQL window functions for Z-score analysis, rolling average spike detection, and IQR outlier detection
  • Flags 7,512 invoices (15%) as suspicious, identifies 35 HIGH-risk vendors out of 210
  • Interactive Tableau dashboard — 🔗 View Live
  • Stack: Python · PostgreSQL · SQL · Pandas · Tableau

🏦 Credit Risk Analytics

End-to-end credit risk analysis on 307,511 real loan applicants — identifying default patterns across demographics, bureau activity, and credit behavior

  • Full pipeline: Python EDA + feature engineering → PostgreSQL business analysis → Tableau dashboard
  • Male applicants default at 45% higher rate than females — under-30s are highest risk segment at 11.47%
  • Bureau activity is the strongest default predictor — 41+ records show 1.7x higher default rate than low-activity customers
  • 4 SQL analysis layers: KPI generation, demographic analysis, credit behavior, advanced risk scoring
  • Stack: Python · PostgreSQL · SQL · Tableau | Live Dashboard →

🚆 Rail Delay Tracker

Automated data pipeline simulating real-time railway delays with risk classification

  • APScheduler runs the pipeline every 5 minutes automatically
  • Classifies delays into HIGH / MEDIUM / LOW risk tiers per train
  • Dual storage — rolling CSV (last 10 records/train) + optional PostgreSQL
  • Interactive Power BI dashboard with delay trends and risk distribution
  • Stack: Python · Pandas · APScheduler · PostgreSQL · SQLAlchemy · Power BI

🌐 Connect with Me

⭐️ From Saksham3124

Pinned Loading

  1. gst-invoice-anomaly-detection gst-invoice-anomaly-detection Public

    3-layer GST invoice anomaly detection system using Python, PostgreSQL, and SQL window functions — with vendor risk scoring and Tableau dashboard

    Python 1

  2. credit-risk-analytics credit-risk-analytics Public

    End-to-end credit risk analysis on 307K loan applicants using SQL, Python & Tableau — identifying default patterns across demographics, age groups, and bureau activity

    Jupyter Notebook

  3. Rail-Delay-Tracker Rail-Delay-Tracker Public

    Real-time railway delay simulation pipeline with automated scheduling, risk classification, CSV + PostgreSQL dual storage, and a Power BI dashboard for trend monitoring and analysis.

    Python 1

  4. Spending-Analytics-Forecasting Spending-Analytics-Forecasting Public

    ML-powered spending analytics system with category prediction, budget monitoring and forecasting

    Python 1

  5. olist-ecommerce-delivery-analysis olist-ecommerce-delivery-analysis Public

    End-to-end SQL analysis of 100K+ Olist orders identifying delivery failures, geographic risk patterns, and seller accountability gaps — visualized in Tableau.

    1

  6. Superstore-KPI-Dashboard Superstore-KPI-Dashboard Public

    Business KPI dashboard analyzing $2.3M Superstore sales using Python, SQL, and Power BI

    Python 1