Skip to content
View akash-pathak-scientist's full-sized avatar

Block or report akash-pathak-scientist

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.

Content in all repositories owned by your account will be closed.
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

Typing SVG

LinkedIn GitHub Portfolio Email



🧠 About Me

class AkashPathak:
    role       = "Data Analyst"
    location   = "India 🇮🇳"
    focus      = ["FinTech Analytics", "Fraud Detection", "Business Intelligence", "Blockchain"]
    stack      = ["SQL", "Python", "Power BI", "Scikit-learn", "Excel", "Tableau", "Flask"]
    currently  = "6 end-to-end analytics projects — ALL COMPLETE ✅"
    goal       = "Data Analyst role @ high-impact BFSI / Product company"
    superpower = "I frame the business problem BEFORE touching the data 🎯"


🚀 Portfolio Projects

🎉 ALL 6 PROJECTS COMPLETE & LIVE ON GITHUB

🏆 Project 01 — FinTech Transaction Analytics

Status SQL Python PowerBI Flask

End-to-end pipeline on 15,000 transactions — KPIs, fraud signals, RFM segmentation, MoM trends, and a live Flask dashboard with 12 REST API endpoints.

Key Results:

  • 🔴 Identified ₹1.45Cr at-risk volume via 3 fraud signals
  • 👥 RFM segmented 2,982 users → Champion · Loyal · At Risk · Dormant
  • 📈 30+ SQL queries — Window Functions, CTEs, Cohort Retention
  • 📊 9-panel dark dashboard + 12 REST endpoints · all 200 OK

View Repo

🛡️ Project 02 — Credit Card Fraud Detection

Status ML AUC Flask

ML pipeline on 50,000 transactions — extreme class imbalance (0.174% fraud). 4 models trained, benchmarked, and served via live animated dashboard.

Key Results:

  • 🤖 Random Forest → 98.14% AUC-ROC (best model)
  • ⚖️ Handled imbalance via class_weight="balanced"
  • 🔍 Feature engineering: log(Amount), Amt_Z, Hour
  • 🚨 3 rule-based SQL fraud detection queries in production

View Repo

🛒 Project 03 — Customer & Sales Analytics

Status KMeans Cohort Flask

Full customer analytics on 50K customers · 200K orders · ₹69.7Cr revenue — RFM segmentation, cohort retention, LTV modeling, and churn prediction.

Key Results:

  • 👥 KMeans RFM → Champion · Loyal · At Risk · Dormant
  • 🔄 12-cohort × 6-month retention heatmap built
  • 💰 LTV analysis by segment, city, channel & age group
  • 📊 10 dark-theme charts · 12 API endpoints · all 200 OK

View Repo

📊 Project 04 — Finance / MIS Dashboard

Status PnL EBITDA Flask

CFO-level MIS across 24 months · 8 departments — P&L waterfall, EBITDA bridge, budget variance RAG status, cash flow & YoY comparison.

Key Results:

  • 💰 FY2024 Revenue ₹13.07Cr | YoY +16.9% growth
  • ⚡ EBITDA ₹1.71Cr (13.1% margin) | PAT ₹1.03Cr
  • 🎯 Budget vs Actual — RAG (Red/Amber/Green) for 8 depts
  • 🏛️ P&L waterfall + EBITDA bridge + 12 financial charts

View Repo

₿ Project 05 — Crypto & Payment Analytics

Status Chains Volume Flask

On-chain analytics across 6 blockchains · 500K transactions · $2.33B volume — OHLCV price action, whale detection, wallet segmentation, DeFi TVL & cross-chain correlation.

Key Results:

  • ₿ 6-chain OHLCV price simulation + BTC candlestick chart (365 days)
  • 🐋 1,500 whale txns detected — disproportionate volume control identified
  • 🔗 Cross-chain Pearson correlation matrix across all 6 networks
  • 🏦 8 DeFi protocols tracked — $25B+ TVL ecosystem analysis

View Repo

🚀 Project 06 — CareerPilot AI

Status Python AI SPA

End-to-end AI job-search platform — NOT just a resume generator. Resume import → AI analysis (7 scores) → ATS optimization → job matching (1-100) → cover letter generation → application tracking.

Key Results:

  • 🔍 AI resume analysis with ATS, Skills, Quantification, Keyword scores + fix recommendations
  • 🎯 Job matching engine 1-100 across LinkedIn · Naukri · Indeed
  • ✍️ Company-specific cover letters from resume only — no fabrication
  • 📋 Full application tracker: Saved → Applied → Interview → Offer → CSV export

View Repo Live Demo



🛠️ Tech Stack

Languages & Query

Python SQL PostgreSQL

Data & ML

Pandas NumPy Scikit--learn Matplotlib

Visualization & BI

Power BI Tableau Excel

Tools & Deployment

Git Flask Jupyter VS Code



📊 GitHub Stats

GitHub Streak



🏅 Certifications

Certificate Issuer Badge
Google Data Analytics Professional Coursera / Google Google
Power BI Data Analyst Associate PL-300 Microsoft Microsoft
SQL — Gold Badge HackerRank HackerRank
Python for Data Science IBM / Coursera IBM


🔥 What Makes Me Different

Skill
🎯 Business-first thinking — I frame the problem before touching the data
🔗 Full-stack analytics — every project: SQL + Python + Dashboard
💬 Stakeholder-ready insights — written for decision-makers, not just data teams
Real datasets · Real queries · Real decisions — no toy examples
🏗️ Production-grade code — Flask APIs, clean SQL, modular Python


💬 Open to Data Analyst roles · Collaborations · Freelance Projects

LinkedIn Portfolio


"Data is the new oil. Analytics is the refinery."

Popular repositories Loading

  1. fintech-transaction-analytics fintech-transaction-analytics Public

    End-to-end FinTech Transaction Analytics | SQL · Python · Power BI · RFM Segmentation · Fraud Detection

    HTML

  2. akash-pathak-scientist akash-pathak-scientist Public

    GitHub Profile README — Akash Pathak | Data Analyst

  3. credit-card-fraud-detection credit-card-fraud-detection Public

    ML pipeline: Credit Card Fraud Detection | Python · Scikit-learn · Random Forest · 98.14% AUC-ROC

    HTML

  4. akash-pathak-scientist.github.io akash-pathak-scientist.github.io Public

    Akash Pathak — Data Analytics Portfolio | SQL · Python · Power BI · ML

    HTML

  5. customer-sales-analytics customer-sales-analytics Public

    Customer & Sales Analytics | 50K Customers · 200K Orders · RFM · Cohort Retention · LTV · KMeans · SQL · Python · Flask

    HTML

  6. finance-mis-dashboard finance-mis-dashboard Public

    Finance MIS Dashboard | P&L · EBITDA · Budget Variance · Cash Flow · 24 Months · 8 Departments | Python · SQL · Power BI · Flask

    HTML