Turning Data into
Actionable Insights
Data Analyst / Engineer
Python • SQL • ETL • Tableau • AWS • Snowflake • dbt • EDA
3+ years building ETL pipelines, executive dashboards, and ML models across SaaS and data-driven environments — translating complex data into strategic business intelligence.
About Me
I'm a Data Analyst / Engineer based in Dublin, Ireland, with an MSc in Data Analytics (Distinction) from National College of Ireland — completed in 2026. I specialise in building end-to-end data solutions — from ETL pipelines and cloud data warehouses to executive dashboards and predictive models that drive real business decisions.
With 3+ years of hands-on experience across the full analytics stack, I've designed scalable ETL workflows using dbt, AWS Glue, and Snowflake; built SaaS KPI dashboards in Tableau; and applied ML models that contributed to a 12% reduction in customer churn. I thrive at the intersection of data engineering, cloud infrastructure, and business intelligence.
Education & Certifications
MSc in Data Analytics ✓ Completed
National College Of Ireland • Distinction · 2025–2026
Bachelor of Computer Applications
Maharaja Agrasen Himalayan Garhwal University • Grade 2:1
IBM IBM Certifications
Work Experience
Professional journey in data analytics
Data Analyst
Technohunk Info Solutions
UP, India
- ▸Built and maintained scalable ETL pipelines using SQL, dbt, and AWS Glue to ingest raw S3 data into Snowflake, reducing manual data preparation time by 20% and improving data quality across 3 business functions.
- ▸Designed 6+ executive-level Tableau dashboards sourced from Snowflake, tracking SaaS KPIs (ARR, churn, retention, feature adoption) for finance, operations, and leadership teams.
- ▸Led large-scale EDA on product and financial datasets using Python, SQL, and AWS Athena, improving reporting efficiency by 25% and driving 3 product roadmap decisions across 5+ product lines.
- ▸Applied ML techniques (Random Forest, Logistic Regression, K-Means) on 12,000+ customer records to build churn prediction and segmentation models — 80% accuracy, contributing to a 12% churn reduction within 3 quarters.
- ▸Optimised SQL queries on Snowflake and Athena, reducing query runtime by 35% and cutting dashboard load times across key reporting workflows.
- ▸Standardised reporting frameworks and data definitions across finance and operations, reducing errors by 40% and saving ~3 hours/week in manual reconciliation.
- ▸Defined 15+ KPIs in partnership with product, finance, and operations stakeholders; managed analytics workflows in Agile/Jira and documented data models in Confluence.
Data Analyst Intern
Technohunk Info Solutions
- ▸Assisted in data cleaning, transformation, and validation of raw datasets using SQL and Excel to support ongoing analytics projects.
- ▸Conducted exploratory data analysis to identify trends and anomalies, supporting senior analysts in insight generation.
- ▸Developed reports and dashboards in Tableau to track key business metrics and performance indicators.
- ▸Wrote and optimised SQL queries to extract and analyse data from relational databases in Agile sprints.
What I Do
Core services I deliver as a Data Analyst / Engineer
Data Analysis & EDA
Deep-dive exploratory analysis on large datasets to uncover patterns, trends, and outliers that drive business understanding. From raw CSV to boardroom insight.
SQL & Database Design
Complex query writing, data modelling, performance optimisation, and scalable analytics-ready pipelines using PostgreSQL, MySQL, and dbt.
Dashboard & Visualisation
Executive-ready interactive dashboards in Tableau, Power BI, and Cognos that turn complex data into clear narratives for non-technical stakeholders.
Machine Learning
End-to-end ML pipelines using XGBoost, Random Forest, and deep learning. Feature engineering, model evaluation, SHAP interpretability, and deployment.
Business Intelligence
KPI design, SaaS metric tracking (ARR, churn, retention), and reporting frameworks that align data outputs with strategic business goals.
Causal & Statistical Analysis
Going beyond correlation — applying SDID, SHAP, LIME, and causal inference methods to validate true relationships and support evidence-based decisions.
ETL Pipeline Engineering
End-to-end ETL design using dbt, AWS Glue, and Snowflake — from raw S3 ingestion to analytics-ready datasets, with data quality checks and transformation logic built for scale.
Cloud Data & AWS
Building cloud-native analytics infrastructure on AWS — S3, Athena, Glue, Redshift, RDS, CloudWatch — optimised for cost, performance, and reliability at scale.
Featured Projects
Real-world data analytics case studies — click any card to explore in depth
View all repositories on GitHubDiabetes Readmission Prediction
Logistic Regression & ML pipeline on 101,763 hospital records to predict high-risk patient readmissions. Academic research at NCI.
E-Commerce Sales Analytics
RFM clustering + Random Forest on 541,909 UCI transactions. R²=0.98, CRISP-DM methodology, customer segmentation.
HR Dashboard Analytics
Tableau dashboard revealing 16.12% attrition rate across 1,470 employees. Sales dept = 56% of total turnover. Embedded live.
Football Match Prediction
XGBoost on 18,894 matches across 10 European leagues. SHAP + LIME interpretability. 150+ engineered features.
Causality in Video Recommendations
SDID causal validation on KuaiRand-1K (3.13M interactions). 54.2% precision gain over standard DiD. NCI MSc research.
Road Collisions Analysis
XGBoost & Random Forest on 104,258 UK collision records. Speed limit top predictor. CRISP-DM. SMOTE balancing.
EV Adoption Impact Analysis
EPA + DOE data pipeline using MongoDB & PostgreSQL. King County: 200k+ EV registrations. AQI vs. adoption correlation.
Mining Logistics Simulation
Discrete event simulation using SimPy. 94% throughput prediction accuracy. Fleet sizing optimisation via what-if analysis.
Multimodal Emotion Recognition
ResNet-50 + BERT fusion for 7-class emotion classification from images and text. SMOTE for class imbalance handling.
Bank Loan Report
Interactive Tableau dashboard analysing loan portfolio health — good vs. bad loan KPIs, funded amounts, repayment trends, and borrower risk profiles.
HR Analytics Dashboard
Explore the live Tableau dashboard analysing attrition patterns across 1,470 employees. Interact with filters, drill down by department, and see real workforce insights.
Bank Loan Report
Explore the live Tableau dashboard analysing loan portfolio health — track good vs. bad loan KPIs, funded amounts, repayment trends, and borrower risk profiles across the full dataset.
Technical Skills
Tools and technologies across the full analytics & engineering stack
Hover any node to reveal proficiency · 3 orbital rings · 18 technologies
Also Familiar With
Full stack of tools across the analytics & engineering ecosystem
Data Engineering
Cloud & AWS
Analytics & BI
Programming & Libraries
Databases
Automation & Monitoring
Data Insights
Thoughts on data analytics, machine learning, and business intelligence
April 2025 · 6 min read
Why XGBoost Outperforms Deep Learning on Tabular Data
A breakdown of why gradient boosting consistently beats neural networks on structured tabular datasets — and what the evidence from football match prediction and causal ML reveals about choosing the right model for the job.
Vishal Chaudhary
Coming Soon
March 2025 · 8 min read
From Raw SQL to Executive Dashboard: A Data Analyst's Pipeline
The full journey from messy data to a polished executive dashboard — covering SQL modelling, dbt transformations, Python cleaning, and Tableau visualisation. With real examples from SaaS KPI tracking.
Vishal Chaudhary
Coming Soon
February 2025 · 7 min read
When Correlation Lies: Lessons from Causal ML in Recommendation Systems
How standard ML models in video platforms confuse correlation with causation — and why Synthetic Difference-in-Differences (SDID) delivered 54.2% better causal precision than traditional methods.
Vishal Chaudhary
Coming Soon
Words on Data
Wisdom that shapes how I think about analytics
Without data, you're just another person with an opinion.
Data is the new oil. It's valuable, but if unrefined it cannot really be used.
Torture the data, and it will confess to anything. The goal is insight, not confirmation.
Always Learning, Always Growing
Let's Connect
I'm actively seeking data analytics roles in Dublin and open to remote opportunities. Whether it's a full-time role, contract, or collaboration — let's talk.
Typically responds within 24 hours.