About Skills Projects Education Certs Contact
Open to opportunities

Bandi Charishma

📋 hire_charishma.sql
SELECT BEST_ANALYST()
FROM candidates
WHERE name = 'Charishma'
  AND gpa > 8.9
  AND skills IN ('SQL','Python','PowerBI')
  AND vibes = 'immaculate';
✓ 1 row returned — HIRE HER IMMEDIATELY
-- p-value: 0.0001 (statistically obvious)
Data Analyst who turns raw numbers into real decisions

Turning raw data into decisions.
Building dashboards that tell the story behind every number.
SQL · Python · Power BI · Tableau · Excel

✦ Hire Me ↓ View Projects ↓ Resume
0
Projects Built
← real impact!
0
$M Data Analyzed
← loan portfolio
0
K+ Records Processed
← real datasets
8.9
GPA / 10.0
← top of class!
📊 proficiency.csv
Excel / Power Query90%
SQL88%
Power BI + DAX85%
Python (Pandas)80%
Tableau75%
self-assessed, project-verified ✓
🧪
Data Nerd Confession: I check if data is normal before trusting a mean. Shapiro–Wilk test? Yes, always.
H₀: μ₁ = μ₂ → p = 0.002 → REJECT ✓
skills
■ BI Tools 50%
■ SQL/Python 30%
■ Stats/ML 20%
who i am

Analyst by training. Problem-solver by nature.

I find the story inside the chaos of raw data — and tell it in a way that makes people act.

I'm Bandi Charishma — a final-year Computer Science student at GRIET Hyderabad with a deep obsession for finding the story inside raw data. I don't just build dashboards; I build tools that help people make better decisions, faster.

My toolkit spans SQL for deep querying, Python for analytical workflows, and Power BI & Tableau for turning numbers into visual narratives. I've processed over 98,000+ real records, built KPI frameworks, and delivered insights that simulate real business impact — all before my first full-time role.

What drives me is the moment a chart makes someone say, "I never saw it that way before." That's when data does its real job.

this is the part where correlation ≠ causation,
but curiosity + rigor always = insights
📓 field_notes.py
def charishma():
  passion  = "data storytelling"
  goal    = "first analyst role in 2025"
  belief  = "noise → signal → decision"
  return insights, impact, growth
← look here! 📌
🔍 Fun data fact about me:
If you run a chi-square test on my work ethic vs. results,
you'd get χ² = 47.3, p < 0.001 — statistically significant! 😄
Translation: results are no coincidence.
📍
Location
Hyderabad, Telangana, India
🎓
Education
B.Tech CSE · GRIET · 2022–2026 · GPA 8.9
✉️
Email
charishmareddy012@gmail.com
💼
Status
● Open to full-time roles
🐙
📈 skills vs. time — trending up!
↑ growth
technical arsenal

The tools I speak fluently.

Hover the tags — each one has a secret annotation. Because data people notice details. 👀

since 2023 ✍️
🗄️
Data Querying & Analysis
SQLJOINs, CTEs, Window Fns JOINsLEFT JOIN is my love language Window FnsLAG · LEAD · RANK · DENSE_RANK CTEsWITH clause = clean queries Subqueriesnested selects for complex logic GROUP BYaggregation is my superpower
SQL88%
Excel / Power Query90%
fav tool! 💚
📊
Business Intelligence & Viz
Power BIDAX + Time Intelligence = 🔥 DAXCALCULATE · SUMX · RELATED TableauLOD calcs change everything KPI DashboardsMTD · MoM · YoY tracked Slicersinteractive filtering = executive happy
Power BI + DAX85%
Tableau75%
🐍 snake charmer
🐍
Programming & Analytics
Pythondf.groupby().agg() all day Pandasthe real MVP of analytics NumPyvectorized ops = fast Rlm() · ggplot2 · tidyverse Matplotlibplt.show() → ahhh beautiful
Python (Pandas/NumPy)80%
R (Statistical)65%
this is where 🧠 lives
🧠
Analytical Concepts
EDAALWAYS plot before modeling (Anscombe '73) Data CleaningNULL ≠ zero — huge difference! Data Modelingstar schema · 3NF · slowly changing dims Statisticst-test · chi² · ANOVA · regression ML BasicsGA+NN hybrid for bankruptcy prediction
EDA & Data Cleaning87%
Statistical Analysis78%
Remember: correlation ≠ causation — but a good analyst knows the difference 🔍
r(x,y) = 0.94 → wow, strong!
but does x cause y?
→ plot it. question it. verify it.
r=0.94
portfolio

Where data meets decisions.

Real datasets, real insights, concrete business impact — every project tells a story.

// Customer Intelligence
🔗⚙
highest!
Customer Segmentation Dashboard
Identified which customers actually drive revenue. Built a full segmentation pipeline using Excel Power Query — cleaned messy raw data, engineered behavioral features, and surfaced income bands, loyalty tiers, and spending clusters.
📌 key finding
Top 20% of customers → 68% of revenue (Pareto confirmed). Segment B shows 3.2× higher LTV vs. baseline.
4
Segments
↑32%
Targeting
ExcelPower QueryPivot TablesEDA
// Financial Risk Intelligence
🔗⚙
risky zone!
Financial Loan Analytics Dashboard with KPI Tracking
Analyzed a $435M loan portfolio spanning 38,000+ applications. Built automated MTD and MoM KPI tracking — transforming manual monthly reporting into a real-time monitoring system.
📌 key finding
13.8% high-risk loan concentration identified → triggered policy review. MoM default rate shift of +2.1% flagged early.
38K+
Applications
$435M
Portfolio
13.8%
Risk Flagged
Power BIDAXKPI TrackingRisk Analysis
// Ride Operations Analytics
🔗⚙
↑ peak!
Ola Ride Data Analysis & Business Insights Dashboard
Rebuilt ride operations visibility for 60,000+ Ola bookings across ₹16M in revenue. Slicer-driven Power BI dashboard with real-time KPI refresh for bookings, cancellations, and driver ratings.
📌 key finding
Peak demand: Friday 6–9 PM, yet driver availability lowest. Cancellation rate 3× higher in Zone B → geo-targeting recommended.
60K+
Bookings
↓50%
Report Time
₹16M
Revenue
Power BIDAXOperations AnalyticsSlicers
// Academic · ML Research
📄 IEEE ⚙
★ Published in IEEE — An Evolutionary Computing Model for Financial Distress Prediction View on IEEE Xplore ↗
bankrupt?
Financial Risk Analysis & Bankruptcy Prediction (ML)
Designed a hybrid AI system fusing Genetic Algorithms and Neural Networks to predict corporate bankruptcy. GA for intelligent feature selection on imbalanced datasets, NN hyperparameter optimization.
📌 research insight
GA-selected features reduced dimensionality by 62% while improving F1. Precision-recall tradeoff optimized for min false negatives.
GA+NN
Hybrid Model
↑F1
Imbalanced Data
PythonGenetic AlgorithmNeural Networksscikit-learn
🏛️
Published · IEEE Xplore
An Evolutionary Computing Model for Financial Distress Prediction
↗
academic journey

Built on a strong foundation.

2022 — 2026
B.Tech in Computer Science & Engineering
Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad
✦ GPA: 8.9 / 10.0

Focused on data structures, DBMS, OOP, operating systems, and machine learning — building the CS depth that makes me a stronger analyst than most.

2020 — 2022
Intermediate (MPC)
Sri Chaitanya Junior College, Hyderabad
✦ Score: 96.9%
credentials

Certified. Validated. Verified.

Industry certifications from globally recognized platforms — credibility beyond the classroom.

✓
HackerRank
SQL Certification
July 2025
JOINs, aggregations, window functions
✓
Newton School
Data Analytics using Power BI
March 2025
DAX, KPIs, interactive dashboard design
✓
Cisco Networking Academy
Data Analytics Essentials
August 2025
Data cleaning, visualization, real-world datasets
✓
Deloitte
Data Analytics Job Simulation
June 2025
Client-ready insights, business data analysis
🥚 Data Nerd Easter Eggs
because real analysts hide (and find) details in the data
📐
Anscombe's Quartet
4 datasets. Same mean, variance, correlation (r=0.816). Completely different shapes.
ALWAYS. PLOT. YOUR. DATA. 🎨
📉
Correlation ≠ Causation
Nicolas Cage films correlate r=0.97 with pool drownings. Doesn't mean Nic Cage causes drownings.
r(x,y) ≈ 1 ≠ x causes y 🚫
🎲
Simpson's Paradox
A trend in subgroups disappears or reverses when groups are combined. Always check subgroups!
GROUP BY everything first 🔍
🎯
p < 0.05 Problem
Test 20 random things → one will be "significant" by chance. That's math, not discovery.
H₀ rejected ≠ discovery ⚠️
🍕
Goodhart's Law
"When a measure becomes a target, it ceases to be a good measure." Every KPI person must know this.
metric ≠ goal, goal ≠ metric 🎭
🦆
Survivorship Bias
WWII planes: reinforce where bullet holes AREN'T — the planes that got hit there didn't return.
missing data tells stories too 💡
🧮
Bessel's Correction
Why we divide by n-1 for sample variance. Degrees of freedom matter more than most think.
σ² = Σ(x-x̄)² / (n-1) 📐
scroll through the nerd lore →
🕹️ Konami Code — press in order:
↑
↑
↓
↓
←
→
←
→
B
A
Press all 10 → 🎉 SELECT * FROM candidates WHERE name='Charishma' AND status='hired'; → 1 row found! ✓
let's connect

Ready to contribute from day one.

I'm actively looking for my first full-time Data Analyst role. If you're building a data team that values curiosity, rigor, and clean visuals — let's talk.

✉️
📍
Location
Hyderabad, Telangana · Open to remote
response time: <24 hrs (statistically verified 📊)
n=47 emails, μ=6.2 hrs, σ=3.1
✅ Message sent successfully! I'll respond within 24 hrs 📊
❌ Something went wrong. Please email me directly at charishmareddy012@gmail.com