M.S. data science · rmit ’25 · 3y @ cognizant

Animesh
Dahiya

Currently
—Data Scientist
01 / 04·auto-cycle
Summary

Data science master’s graduate from RMIT with three years as a Software Engineer at Cognizant. I turn messy, real-world data into systems that ship — ML pipelines, analytics platforms, and automation that survive contact with production.

See selected workDownload resumeOr get in touch →
BasedMelbourne AEST +10
Experience3y eng · MS data sci ’25
FocusML / analytics / automation
StatusOpen to work
Reachdahiyaanimesh@gmail.com
Featured — 01 / Pitwall

An F1 analytics platform that predicts the race before lights out.

Read the case study→
pitwall.live · session telemetry
MELBOURNE GP · LAP 34 / 58
Races analysed
0
Lap records
0K+
Model accuracy
0%
FastAPISQLiteXGBoostSHAPReact/TSTailwind
01 — about

Asoftwareengineerwholearnedtolistentodata—andshipit.

I grew up in software engineering before moving into data science. Three years at Cognizant taught me how production systems actually break — so when I moved into ML, I kept the same instinct for shipping things that survive contact with real users.

At Cognizant I architected end-to-end automation using UiPath, Amelia chatbot, and Eyeshare, driving $300K in annual savings — recognised by APAC leadership. I cut ticket-resolution time by 25%, reduced manual effort by 15%+ per month, and ran 100+ workflows across daily, weekly, and ad-hoc schedules.

With a Master’s in Data Science from RMIT (2025), I build the same way: a Python/FastAPI backend sitting on 400K lap records, a Random Forest + XGBoost ensemble across 29 features, and a React/TypeScript frontend that explains its own predictions with SHAP waterfalls. Engineering rigour, data-science tools.

01
0+
Years engineering
Cognizant, APAC delivery
02
$0K
Annual cost savings
Automation, recognised by APAC leadership
03
0+
Automated workflows
Managed under DevOps / Agile
04
0M
Reviews analysed
Employee sentiment study, VADER + TextBlob
02 — stack

ThetoolsIreachfor, arrangedbylayer.

I care about fluency, not a checklist. These are the things I pick up without thinking — split by where they sit in a system.

layer 010tools
Languages & core
Python◦
R◦
SQL◦
TypeScript◦
JavaScript◦
HTML / CSS◦
Java◦
Kotlin◦
Python◦
R◦
SQL◦
TypeScript◦
JavaScript◦
HTML / CSS◦
Java◦
Kotlin◦
layer 020tools
Frameworks & UI
React◦
Next.js◦
FastAPI◦
Flask◦
Node.js◦
Tailwind CSS◦
Streamlit◦
Shiny◦
React◦
Next.js◦
FastAPI◦
Flask◦
Node.js◦
Tailwind CSS◦
Streamlit◦
Shiny◦
layer 030tools
ML / AI
scikit-learn◦
TensorFlow◦
XGBoost◦
Random Forest◦
SHAP◦
Pandas◦
NumPy◦
NLTK / spaCy◦
OpenCV◦
GPT-4 / DALL·E◦
LangChain◦
VADER◦
TextBlob◦
NetworkX◦
scikit-learn◦
TensorFlow◦
XGBoost◦
Random Forest◦
SHAP◦
Pandas◦
NumPy◦
NLTK / spaCy◦
OpenCV◦
GPT-4 / DALL·E◦
LangChain◦
VADER◦
TextBlob◦
NetworkX◦
layer 040tools
Data, cloud & ops
MySQL◦
MongoDB◦
SQLite◦
ETL◦
Docker◦
Git / GitHub◦
GitHub Actions◦
CI/CD◦
AWS◦
Vercel◦
Render◦
MySQL◦
MongoDB◦
SQLite◦
ETL◦
Docker◦
Git / GitHub◦
GitHub Actions◦
CI/CD◦
AWS◦
Vercel◦
Render◦
layer 050tools
Visualisation & automation
Tableau◦
Power BI◦
Jupyter◦
Matplotlib / Plotly◦
ggplot2◦
Folium◦
QGIS◦
UiPath◦
Amelia AI◦
Eyeshare◦
Tableau◦
Power BI◦
Jupyter◦
Matplotlib / Plotly◦
ggplot2◦
Folium◦
QGIS◦
UiPath◦
Amelia AI◦
Eyeshare◦
03 — selected work

ThingsI'veshipped.

01
case study
2025 · ML

Pitwall

Full-stack Formula 1 analytics platform. An end-to-end race-outcome engine built from raw FastF1 telemetry to interpretable predictions.

PythonFastAPISQLiteReactTypeScriptTailwindXGBoostRandom ForestSHAPRenderVercel
Visit live siteView repository
Lap records
400K+
Races
88
Features
29
Accuracy
85%
model: rf + xgb · 29 featuresshap waterfall
  1. 01

    Built a production-grade F1 analytics platform with a FastAPI backend, SQLite database of 400K+ lap records across 88 races (2021–2024), and a React/TypeScript/Tailwind frontend deployed on Render and Vercel.

  2. 02

    Engineered a race-outcome prediction module using a Random Forest and XGBoost ensemble across 29 features with SHAP waterfall explanations for interpretability, reaching 85% prediction accuracy.

  3. 03

    Developed a Pit Stop Strategy Optimizer with tyre-degradation modelling, pit-window analysis, and lap-by-lap race replay; built a FastAPI proxy to resolve CORS for live circuit-map integration.

  4. 04

    Shipped 5+ analytics modules including driver performance dashboards (season arc, teammate battle, pace profiling), a Gantt-style tyre stint map, and a safety-car impact heatmap from track-status data.

022025 · AI

Konzepta

AI-assisted design ideation platform

3–5s
Avg response
5 UX designers
Testers
Next.jsReactFlaskGPT-4.1DALL·E 3+2
GitHub →
032024 · Data

HomeHop

Property value analysis with geospatial ML

973
Properties
87%
XGBoost accuracy
PythonXGBoostDecision TreesStreamlitPower BI+2
042023 · NLP

Employee Sentiment at Scale

NLP + network analysis over 2.3M reviews

2.3M+
Reviews
PythonVADERTextBlobNetworkXGephi
052023 · Viz

Gaming Analytics Dashboard

R Shiny exploration of game sales

RShinyggplot2plotly
Live →
062024 · Tool

YouTube → Spotify

Cross-platform playlist converter

90%+
Transfer rate
FlaskYouTube APISpotify APIOAuth
GitHub →
04 — experience

Realwork,realimpact—shippedtoproduction.

  1. Aug 2023 — Present
    Melbourne, Australia
    Current

    Department Manager

    · McDonald's
    • Managing daily operations and leading a team of 10+ crew members; concurrently completed a Master's in Data Science at RMIT (2023–2025) — multitasking and leadership under real operational pressure.
    OperationsPeopleLeadership
  2. Aug 2020 — Jun 2023
    Pune, India

    Software Engineer

    · Cognizant Technology Solutions
    • Architected and deployed end-to-end automation using UiPath, Amelia chatbot, and Eyeshare — driving $300K in annual cost savings, recognized by APAC leadership for the impact.
    • Reduced manual operational effort by 15%+ per month by deploying the Amelia AI chatbot to automate the incident-ticketing pipeline across teams.
    • Streamlined IT operations with Eyeshare: automated database queries, shell scripts, and ServiceNow incident workflows — cutting resolution time by 25%.
    • Managed and monitored 100+ automated workflows in a DevOps/Agile environment, ensuring zero-downtime execution across daily, weekly, and ad-hoc schedules.
    UiPathAmeliaEyeshareServiceNowPythonDevOpsAgile
  3. Feb 2020 — Jul 2020
    Chennai, India

    Software Engineering Intern

    · Cognizant Technology Solutions
    • Completed a 6-month full-stack engineering program covering Java, Spring Boot, HTML/CSS, and databases as part of Cognizant's graduate training.
    • Gained hands-on exposure to enterprise software practices — version control, code review, and Agile workflows.
    JavaSpring BootHTML/CSSSQL
05 — education

Formaltraining,theneverythingelse.

Jul 2023 — Jun 2025Melbourne, Australia

Master of Data Science

RMIT University

Machine learning, NLP, statistical modelling, and data visualisation.

Jul 2016 — May 2020Chennai, India

Bachelor of Technology

SRM Institute of Science and Technology

Computer Science & Engineering.

publication
2020

A Smart Approach For Securing Lottery System With Blockchain

International Journal of Advanced Science and Technology (IJAST)

Vol. 29, No. 08 (2020), pp. 4009–4015

Ethereum-based payable smart contract combined with a ring signature scheme to make lottery systems transparent, fair, and free of middlemen.

06 — contact

Let’sbuildsomething.

I reply within 48h.