-- data analyst · troy, michigan

Pranita
Tashildar

I turn messy operational data into reporting that leaders trust, with SQL, Python, Power BI and Tableau. Automotive, consulting and construction, since 2020.

pranitat1223@gmail.com LinkedIn GitHub Resume Query my resume
PT
MS Data Analytics
Walsh College, 2025Open to work
Executive reportingdashboards global leadership ran their weekly reviews onStellantis
Real-time monitoringGrafana and InfluxDB incident dashboards with automated alertsMethodica
ForecastingPython and SQL models client planning teams used to set their plansBright Consultancy
Supplier pricingvalue optimization reports surfacing savings from SAP MM pricing dataStellantis
BI from scratchMySQL and Power BI analytics built for construction operationsShree Samarth
Only have 5 minutes?
1  resume PDF 2  flagship demo: fraud detection 3  query my resume in SQL 4  read the build log

Experience

Five roles, three industries, one habit: check every number back to source.

Jan – Feb 2026Auburn Hills, MI

Business Intelligence & Data Analyst

Stellantis (Contract via Methodica Technologies)

Contract BI analyst supporting global Propulsion Systems leadership across international operations.

  • Leadership ran weekly reviews off scattered SAP extracts and Excel files. Built the executive Power BI dashboards that joined them into one view of operational KPIs and project status, and that view became the standing basis for the weekly review.
  • Supplier spend sat in SAP MM pricing data that nobody was comparing across suppliers. Developed the supplier value optimization reports that put those prices side by side, surfacing savings across the supply base.
  • Program slips used to surface at the deadline. Tracked program KPIs, milestones and trends weekly, so leadership saw a slip while there was still time to do something about it.
  • Global engineering and program stakeholders each wanted something different from the same data. Gathered the requirements and turned them into report specs, so reporting got built once instead of reworked per team.
  • The recurring Excel reports were assembled by hand every week, eating analyst time and inviting copy-paste errors. Automated them with validation checks built in, cutting production time while keeping the numbers accurate.

Jan – Apr 2025Troy, MI

Data Analyst

Methodica Technologies

Analyst on a digital-simulation platform delivery, owning data validation and predictive monitoring.

  • Platform incidents were being caught after the fact. Built Grafana dashboards on InfluxDB with automated alert workflows, so the team saw failures in real time instead of hearing about them later.
  • Ambiguous requirements were driving rework across the delivery. Documented them in Visure ALM and built full traceability between requirement, test and result, which tightened scope and cut the back-and-forth.
  • Leadership reporting was only as good as the data feeding it. Owned that gate: SQL (BigQuery), Python and Excel validation ahead of every report, plus the written playbooks the project was later audited against. When anomalies did surface, troubleshot them to root cause and tracked the fixes in JIRA.
  • Automated test runs were generating far more data than anyone was reviewing. Wrote the 13 dashboard requirements and built the Grafana views behind them, so execution health, errors by exception type and system utilization read from one screen.

Oct 2024 – Feb 2025Troy, MI

Graduate Teaching Assistant, Data Analytics

Walsh College

Teaching assistant for the graduate data-analytics program.

  • Graduate students could run the tools but their analyses often stopped at the output. Coached cohorts through hands-on Python, R, Power BI and Tableau work, pushing each project to a conclusion someone could act on.
  • The gap that held most students back was communication, not code. Spent the bulk of the coaching on explaining an analysis to a non-technical audience, the skill that decides whether the work gets used.
  • The role ran across the fall and winter terms alongside a full MS course load: often teaching a method the same week it came due in coursework, which sharpened both sides of the desk.
  • Dashboard critique was the most common ask. Reviewed student Power BI and Tableau builds and showed how chart choice and layout change what a stakeholder takes away from the same numbers.

Jan – Aug 2024Pune, India

Data Analyst

Bright Consultancy Services

Consulting analyst running several client engagements in parallel.

  • Clients needed forward projections they could plan against with confidence. Built the Python and SQL forecasting models that filled that gap, and client planning teams used them to set their plans.
  • Client planning cycles dragged because the reporting behind them was slow to assemble. Designed star-schema Power BI dashboards with DAX measures that put the numbers in front of planners faster, shortening the cycle.
  • Running several engagements in parallel left no room for manual refreshes. Standardized the recurring reporting pipelines with scheduled refresh, so every client got current numbers without analyst hours behind each report.
  • Parallel engagements multiplied the chances for a bad number to reach a client. Built validation rules into every scheduled refresh, so errors were caught in the pipeline instead of in a client meeting.

Jan 2020 – Dec 2023Belagavi, India

Data Analyst, Construction Operations

Shree Samarth Constructions (family business)

First analytics hire for the family construction business, building the function from zero.

  • The business tracked multi-project construction on paper and phone calls. Designed the MySQL databases and the intake routine that turned daily site updates into structured, queryable data.
  • Numbers in a database still were not decisions. Built the Power BI dashboards for resource, budget and progress tracking on top of it, and management started planning from them.
  • Cost overruns used to be found after the money was spent. Compared budget against actual spend by project, flagging variances early enough for managers to act on them.
  • Materials were bought reactively, so sites ran out or over-ordered. Implemented consumption forecasting with automated reorder workflows, which cut material waste and kept deliveries on schedule.

Case studies

The question, the method, and what the numbers said.

Office vs remote, mean score
OfficeRemote

QM 600 · Walsh College · 2025 · team of 3

Remote work and productivity

The remote-vs-office debate ran on anecdotes, so the team tested it: NSW remote-work surveys (3,019 responses) plus a 1,000-worker dataset. Remote workers scored higher on productivity and well-being (p < 0.0001), but remote share alone explained little (R² ≈ 0.02), so the call was a structured hybrid model.

PythonExcelt-testsRegression
Run live demo
Order-taking errors by type890 errors · 3,000 orders

QM 602 Lean Six Sigma · Walsh College · 2025

Six Sigma on a pizza line

890 errors in 3,000 orders needed a root cause. X̄–R charts showed crust thickness in control, so the defects were in order-taking: two error types made up 78% of errors at 49,444 DPMO. Fix: address validation and time-stamping at the POS.

SPCParetoDPMODMAIC
Run live demo
Monthly sales per store, $K
BeforeAfter

QM 601 Research Methods · Walsh College · 2025 · team of 3

Do loyalty programs lift retail sales?

A retailer cannot casually A/B test a loyalty program, so the design had to stand on its own. Designed the study (survey, n = 384 via Cochran's formula, paired t-test and regression) and tested it on simulated store data: +20.7% mean monthly sales, p < 0.01.

Research designSamplingPaired t-test
Run live demo
Held-out accuracy by model983 test samples

BE final-year project · Gogte Institute of Technology · 2022 · team of 5

Predicting drinking-water potability

Can nine lab measurements flag unsafe water? Compared eight scikit-learn classifiers on 3,276 samples. Best model (SVM) reached 67% against a 61% baseline; features barely correlate with potability (|r| ≤ 0.04), which caps any model. GitHub

Pythonscikit-learnXGBoost
Run live demo

Client work

Strategy, websites and proposals for a Bloomfield Hills sign company and its owner's music label.

Dempster Designs website homepage

Dempster Designs · Bloomfield Hills, MI · 2026

Website redesign for a 60-year sign company

The company's site no longer matched six decades of work. Built a five-page redesign prototype (HTML, CSS, vanilla JS) with services, story, a 36-photo gallery and contact from the client's own copy and photos, and presented it to the owner.

HTML/CSS/JSNetlify
CodeView prototype ↗
The Blue Guitar website homepage

The Blue Guitar · Dolphin Bay Records · 2026

Artist site prototype for The Blue Guitar

The label needed one place where listeners stream and supervisors license. Built a catalogue prototype with 30-second previews, "listen on" links to the major platforms and a licensing section for film and TV, presented as a client preview.

Web designAudio previews
View prototype ↗

Consulting · 2026

Growth plans for both brands

  • Streaming-first release plan for the label, built on researched 2026 platform pay rates, payout timing and rights registration.
  • Marketing plan for the sign business anchored in an ROI model and break-even analysis, with tiered budgets across local search, social and paid media.
  • One shared social workflow (content calendar, scheduling and a monthly performance scorecard) with separate strategies and voices per brand.
  • Delivered as two written proposals and two decks, presented to the owner.

More projects

Work, internship and coursework.

Methodica Technologies · 2025

Test-results monitoring in Grafana

Test benches produced runs nobody could read at a glance. Wrote the 13 dashboard requirements and built the Grafana views on InfluxDB test-run data, so execution health, errors by exception type and utilization read from one screen.

GrafanaInfluxDBPython
Run live demo

Enerzi Microwave Systems · internship · 2022

IoT sensor data logger

Conditions needed monitoring with no server to run. An ESP8266 and DHT11 stream every reading to Google Sheets, a live remote log at zero infrastructure cost.

IoTArduino C++
Run live demo

IT 599 Capstone · 2025

Explainable fraud detection for digital payments

Fraud flags nobody can explain do not get trusted. The capstone built a hybrid Random Forest, XGBoost and autoencoder framework on 5,000 synthetic transactions with SMOTE, SHAP and LIME; the live demo rebuilds the explainable scoring in your browser.

PythonXGBoostSHAP
Run live demo

QM 505 · 2025

SBA 7(a) loans and job creation

Do loan size and borrower profile drive job creation? Pivoted and regressed 5,000 loan records; the attributes predict almost nothing, and saying so is the finding.

PythonRegression
Run live demo

IT 547 · 2025

Fitness & nutrition database

Fitness data scatters across apps. Designed an ERD-to-SQL relational schema with associative tables, then proved it with indexed analytical queries.

SQLERD
Run live demo

QM 504 · 2024

Equine ulcer treatment analysis

Three ulcer treatments, unclear payoff per dollar. Compared cost against efficacy from cited veterinary sources: ranitidine cost the least per efficacy point; omeprazole reached 85% at 7.8x the cost.

PythonExcel
Run live demo

IT 545 · 2025

House price prediction

Price models overfit as easily as they fit. Compared Linear, Ridge and Lasso regression on the Kaggle House Prices dataset to see what generalizes.

Pythonscikit-learn
Run live demo

IT 544 · 2025

U.S. air quality analysis

One pollution dataset, three audiences. Chose the right chart for each: trends, seasonality and pollutant drivers, so everyone reads the same data correctly.

PythonData viz
Run live demo

IT 546 · 2024

Text analytics: War and Peace

Could an Excel text workflow survive a whole novel? Word frequency and dictionary sentiment, first built in Excel, now runs on the complete text.

ExcelText analytics
Run live demo

AWS Academy · 2022

AWS cloud labs

One server is a single point of failure. Put EC2 web servers behind an Elastic Load Balancer with health checks, so traffic survives an instance dying.

AWS EC2ELB
Run live demo

On GitHub

github.com/Pranita2798

5

data-bi-portfolio

Python to Power BI on public data: EV charging reliability, vehicle recalls, freight and energy forecasting, data-quality audit.

 6

data-analytics-projects

Cohort retention, funnel analysis on BigQuery, A/B test readout, KPI one-pager, SQL library, ETL pipeline.

 6

data-tools-techniques

Segmentation, churn prediction, statistics in R, Tableau, advanced Excel and Grafana monitoring.

Toolkit

Skills, education and certifications.

Query & code
SQLPythonRBigQueryMySQLPostgreSQLInfluxDBSAP (MM pricing)
BI & reporting
Power BI (DAX)TableauGrafanaExcel (Power Pivot, VBA)
Methods
ForecastingHypothesis testingRegressionA/B evaluationSPC & Six Sigma toolsETL & validation
Working style
Stakeholder requirementsRoot-cause analysisData storytellingJIRA & Agile
MS, Data AnalyticsWalsh College, Troy, MI · 2025
BE, Electronics & CommunicationKLS Gogte Institute of Technology, India · 2023
  • Google Analytics Certification (Skillshop)2025
  • Google Analytics IQ & GA4 (Skillshop)2025
  • AWS Academy Cloud Foundations2022

Query this resume

Everything above, as a database. SQLite compiled to WebAssembly, running in your browser.

A data analyst should be queryable. Boot the database and run SQL against tables like experience, accomplishments (situation, action, result), projects and skills. Nothing leaves your browser.

-- let's talk

Open to Data Analyst and BI roles.

On-site, hybrid or remote.

Message on LinkedIn

SELECT * FROM contact;

emailpranitat1223@gmail.com
linkedin/in/pranita-tashildar
github/Pranita2798
resumePranita_Tashildar.pdf
locationTroy, MI (Eastern Time)
build_log/build-log
work_modeon-site / hybrid / remote
statusopen to work

8 rows in 0.04 ms