Data Analyst · Delhi NCR, India

Pratiksha
Dandriyal

SQL · Power BI · Python · DAX

Turning raw data into decisions. I build end-to-end analytics — from data modeling and ETL pipelines to interactive Power BI dashboards — with a focus on answering real business questions, not just building pretty charts. Lately that's extended into building the products themselves, not just the dashboards behind them.

View Projects Download Resume ↓ Get in Touch
Pratiksha Dandriyal, Data Analyst

About me

I build analytics that actually get used

I'm a BTech CSE graduate who builds analytics the way I'd want to receive them: business problem first, clean data pipeline second, and the visualisation last — because a beautiful dashboard built on dirty data is just a pretty lie. At MAX Healthcare, I applied this to 23,000+ patient records across outpatient, ICU, and oncology data, translating clinical business logic into checkable rules and feeding findings into my manager's departmental discussions. I've since built that same discipline into four independent, domain-specific projects — hospital readmission risk, IT helpdesk SLA analytics, SaaS product metrics, and workforce automation risk — each engineered end-to-end from SQL to Power BI, not templated from a Kaggle dataset. What I enjoy most is the debugging trail as much as the finished dashboard — catching a mislabeled measure or a silent SQL bug tells you more about how someone actually works than a clean final chart ever will. That same instinct — real logic over decoration — is why I've also started building outside BI tools: development projects below, built with the same rigor as the dashboards.

Technical skills

What I work with

Analytics & Visualization
Power BIDAXPower QueryKPI TrackingData StorytellingEDAStatistical Analysis
Data & SQL
SQLSQL ServerData ModellingStar Schema DesignETL ConceptsData CleaningRelational DB Design
Programming
PythonPandasNumPyMatplotlibHTML / CSS / JS
Tools & Productivity
MS Excel (Pivot Tables)Git & GitHubVS CodeJupyter NotebookGoogle Sheets
AI & Productivity
AI-Assisted CodingPrompt EngineeringAI-Augmented EDA

Work experience

Where I've worked

Sep 2024 — Dec 2024

EasyGov (Surajya Services)

Internship

Data & Product Analytics Intern

  • Built a weekly reporting pipeline using SQL, Python, and Excel to extract and clean operational data across 5+ departments — delivering structured insights that directly shaped how the product team prioritised resource allocation.
  • Identified recurring system failure patterns and queue time bottlenecks through SQL trend analysis — findings directly incorporated into scheduling workflow changes by the product team.
  • Translated product team requirements into structured weekly reports using Pandas, NumPy, and Matplotlib covering operational throughput, failure rates, and usage trends.

Feb 2024 — Aug 2024

MAX Healthcare · IT Division

Internship

Data Analyst Intern

  • Analysed 6 interconnected Excel files with 23,000+ patient records linked across outpatient, inpatient, ICU, lab, and oncology datasets using common keys to surface cross-departmental patterns.
  • Identified a 38-point oncology margin gap across locations (23.3% at Patparganj vs 61.4% at Alexis) despite Patparganj generating the highest absolute revenue (₹18.6L) — flagging a cost-line review opportunity independent of volume.
  • Analysed ₹2.4 crore in ICU revenue across 188 patients and found Room Rent and Lab Services outweighed Surgery/Procedures — reframing ICU cost forecasting around length-of-stay rather than procedure mix.
  • Contributed to business logic for CLABSI, CAUTI, SSI, and VAP infection tracking — matching billed procedures against lab tests within defined time windows, and helped identify why active (non-discharged) admissions were being missed from infection reports.

Projects

Things I've built

Four end-to-end dashboards across healthcare, IT operations, SaaS, and workforce analytics — each with a domain-specific business narrative, not a generic template.

Project 01

Helpdesk Performance & SLA Analytics

37.5% Network SLA breach 8,000 tickets · 26 months

IT managers can't tell whether missed SLAs are a staffing problem, a routing issue, or category-specific. This dashboard makes it visible and actionable.

SQL ServerPower BIPython
Case Study →

Project 02

SaaS Product Analytics Dashboard

$90K MRR at risk from churn 1 in 4 users retained

A SaaS company loses nearly as much revenue to churn as it earns. This dashboard identifies exactly where — and which fix has the highest ROI.

SQL ServerPower BIPython
Case Study →

Project 03

Hospital Readmission & Patient Flow Analytics

17.5% overall readmission rate 9,093 encounters · 7,500 patients

A synthetic, ICMR-calibrated Indian hospital dataset built ground-up across SQL, Python, and Power BI — cross-validated at every layer to identify which readmission risk factors are real, and which a hospital can actually act on.

SQL ServerPower BIPython
Case Study →

Project 04

AI Job Displacement & Reskilling

41% earned less post-reskilling 13,700+ jobs · 9 countries

Reskilling budgets are allocated by assumption. This dashboard quantifies where AI disruption is concentrated and where intervention is most urgent.

SQL ServerPower BIPython
Case Study →

Development

Also building the product side

Two front-end projects where the same instincts behind the analytics work — real statistical methods, honest scoping, showing the logic rather than hiding it — get applied to building an actual working product instead of a dashboard.

Dev 01

Ledgerline — Expense Tracker

ReactViteRechartsPapaParse

A React expense tracker built around one idea: every insight should be backed by an actual statistical method, not a decorative chart. Categorisation runs on a keyword engine, the Forecast page projects month-end spend using least-squares linear regression, and a Suggestions engine flags outliers by z-score and recurring charges by coefficient of variation. It also handles shared expenses, savings goals, budget pacing, and a printable monthly report — with a dedicated Methodology page documenting exactly how every number is calculated.

  • Forecasting uses a real least-squares regression on the month's daily spend rather than a naive average — the same instinct as catching a bad denominator in a Power BI measure.
  • No backend by design for v1: kept it a zero-setup, client-side demo, with state structured so a database swap later is a contained change, not a rewrite.

One-click demo dataset available — no setup needed to explore.

Live Demo → GitHub ↗

Dev 02

Field Log — Task Tracker

HTMLCSSJavaScript

A task tracker that goes beyond a checklist by layering in real productivity analytics: a day-streak counter, category and priority completion rates, a 12-week GitHub-style activity heatmap, a 14-day trend drawn on canvas with no charting library, and auto-generated insights like "Work is your most consistent category at 82% completion" — all computed from actual stored task data, not decoration.

  • No framework, no bundler, no build step — every function (toggleTask(), computeStreak(), renderAnalytics()) maps directly to a line I can point to and trace in dev tools, with nothing compiled in between.
  • Data persists via localStorage, framed as a scoped v1 choice rather than a gap — multi-device sync through a small backend is the named next step.

One-click demo dataset available — no setup needed to explore.

Live Demo → GitHub ↗

Education & certifications

Formally trained, self-driven

B.Tech — Computer Science & Engineering

Dr. APJ Abdul Kalam Technical University, India

CGPA 8.23 / 10.0 Graduated Sep 2025

Microsoft Power BI Desktop for Business Intelligence

Udemy · 2026

Python for Data Analysis & Business Intelligence

Udemy · 2026

SQL for Data Analysis: Advanced SQL Querying Techniques

Udemy · 2025

Introduction to AI

IBM SkillsBuild · 2024

Let's connect

Open to opportunities

Looking to join a team where I can go beyond building dashboards — owning data pipelines from raw extraction to the final insight.

pratikshadandriyal@gmail.com linkedin.com/in/pratikshadandriyal github.com/pratikshadandriyal
Download Resume ↓