Available for opportunities

David Poudel

Data Analyst · Data Engineer

I build pipelines, models and dashboards that turn messy data into clear decisions. Based in Melbourne — currently studying a Master of Data Analytics at Melbourne Institute of Technology.

Experience & Impact

Data-Driven Thinking.
Human-Centred Approach.

I didn't start in data. I started in curiosity — pulling apart systems, asking why things work the way they do. From my Computer Science degree at Deerwalk Institute of Technology in Nepal to engineering data systems in Melbourne, every step has sharpened how I think about information.

My experience spans backend development, database administration, and data engineering across Cedargate Technologies, HawkAI, CloudFactory, and Deerhold — giving me a full-stack understanding of how data moves and where it breaks.

Today, while completing my Master of Data Analytics at Melbourne Institute of Technology, I'm deepening my focus on machine learning, analytics engineering, and building data systems people actually rely on.

📐
Analytical Rigour
Digging until the numbers make sense.
🔗
Pipeline Thinking
Data flows that stay reliable over time.
📊
Clear Storytelling
Making insights legible to anyone.
🎯
Impact Focused
Success measured by decisions made.

Career Progression

🎓
BSc Computer Science & IT
2017 – 2021
Deerwalk Institute of Technology
💻
Backend Developer Intern
2020 – 2021
Deerhold Private Limited
⚙️
Data Engineer Intern
2022 – 2023
Cedargate Technologies
🗃️
Database Administrator
2023 – 2024
HawkAI
🚀
Master of Data Analytics
2024 – Ongoing
Melbourne Institute of Technology

My Work

ROC Curve — Churn Prediction Model 0.0 0.25 0.5 0.75 1.0 0.0 0.5 1.0 False Positive Rate AUC = 0.836 80% Accuracy Logistic Regression Random baseline Machine Learning
Customer Churn Prediction
Logistic Regression model predicting telecom customer churn with 80% accuracy and 0.836 ROC-AUC. Full EDA, feature engineering, and model comparison across 7,032 records.
Pythonscikit-learnPandasSeaborn
Last Touch vs Shapley Attribution Overcredited Undercredited Email Paid Search Organic Social Display Overcredited by Last Touch Undercredited by Last Touch Analytics
Marketing Attribution Model
Compared Last Touch, First Touch and Shapley value attribution across 10,000 customer journeys. Found Display Ads undercredited and Email overcredited — reshaping $1.3M in revenue attribution.
PythonShapley ValuesPandasMatplotlib
Data Warehouse Architecture SEEDS customers.csv orders.csv products.csv Raw source data STAGING stg_customers stg_orders stg_products Clean + rename Type casting 6 tests passing INTERMEDIATE int_customer _orders Join + aggregate 4 tests passing MARTS customer_lifetime _value CLV + segments 9 tests passing 19 Tests Passed 100% Coverage dbt + DuckDB Data Engineering
Data Warehouse Migration
Built a modern 3-layer warehouse architecture using dbt and DuckDB — staging, intermediate and marts. 19 data quality tests with 100% pass rate and full lineage documentation.
dbtDuckDBSQLYAML
Real-Time Sales Dashboard LIVE Total Revenue $84,320 Total Orders 342 Avg Order $246 Top Store Melbourne Revenue by Store Melbourne CBD $24.1k Sydney CBD $18.3k Brisbane $14.8k Perth $12.1k Adelaide $9.8k Live feed: Laptop Pro · Melbourne CBD · $1,299 · Credit Card · just now Data Engineering
Real-Time Sales Dashboard
Live POS data streaming dashboard built with Streamlit and Plotly. Simulates real-time transactions across 5 Australian stores with KPI cards, interactive charts, and auto-refresh.
PythonStreamlitPlotlySQLite

My Toolkit

Languages & Querying
PythonSQLRBashdbt
Data Engineering
AirflowSparkKafkaStreamlitFivetran
Cloud & Warehouses
BigQuerySnowflakeDuckDBAWSGCP
Analytics & BI
TableauLookerPower BIPandasPlotly
ML & Statistics
scikit-learnPyTorchstatsmodelsA/B Testing
Databases
PostgreSQLMySQLMongoDBSQL Server

Certifications

Introduction to Programming Using Python
Coursera
Core Python programming, data types, control flow, functions and problem solving.
Data Scientist with Python
DataCamp Career Track
Comprehensive track covering data manipulation, visualisation, machine learning, and statistical analysis using Python.
Get In Touch
Let's build something
meaningful.

Always happy to chat about data, pipelines, or new opportunities. Based in Melbourne and open to full-time roles and contracts.