Waleed Sabir

Data Analytics · Data Engineering · Applied Machine Learning

Turning complex data into useful systems. I build analytical pipelines, predictive models, business intelligence solutions, and automated workflows.

Available for opportunities
View Work → Download CV
Stack Python SQL Power BI DAX PostgreSQL
Portfolio Telemetry sys.stats
01 Selected Projects
01 / E-Commerce Intelligence

Olist Marketplace
Analytics Platform

An end-to-end business intelligence system transforming Brazilian e-commerce marketplace data into sales, logistics and product intelligence.

SQL Power BI DAX Star Schema SSAS
Pipeline Architecture SOURCE DATA DATA VALIDATION SQL TRANSFORMATION STAR SCHEMA FACT TABLES DIMENSIONS BI / ANALYTICS SALES LOGISTICS PRODUCTS
0 Orders Analysed Verified · Olist public dataset
16M Revenue (BRL) Verified · 1.5 years of transactions
2016–18 Data Period Observed · Olist public dataset
0 Analytical Areas Sales · logistics · products · customers · payments
FACT_ORDER FACT · 99,441 ROWS DIM_CUSTOMER DIMENSION DIM_PRODUCT DIMENSION DIM_SELLER DIMENSION DIM_DATE DIMENSION FACT_PAYMENT FACT

← Click any table to inspect its structure, grain, row count and purpose.

Interactive Mockup Structural layout only — not a live report — values are illustrative
OLIST E-COMMERCE INTELLIGENCE INTERACTIVE MOCKUP Illustrative layout · Not a live report TOTAL ORDERS 99,441 OBSERVED · 2016-2018 DATA PERIOD 2016-18 OBSERVED · OLIST DATASET DATA DOMAINS 6+ ORDERS · CUSTOMERS · SELLERS ··· ANALYTICAL AREAS 5 SALES · LOGISTICS · PRODUCTS ··· REVENUE VS FREIGHT COST ILLUSTRATIVE · RELATIVE SCALE · NOT ACTUAL VALUES Q3’16 Q4’16 Q1’17 Q2’17 Q3’17 Q4’17 Q1’18 Q2’18 LATE DELIVERY RATE BY STATE (%) ILLUSTRATIVE · SOURCE: OLIST DATASET MARANHÃO (MA) ALAGOAS (AL) CEARÁ (CE) PARÁ (PA) PIAUÍ (PI) Structural mockup · Illustrative values only · Not a live Power BI report
02 / Predictive Analytics & Clustering

New SMI Supermarkets
Customer Intelligence

End-to-end customer segmentation and predictive modelling for a supermarket chain: Bronze, Silver, and Gold clusters drive ROI-optimised targeting strategy.

Cluster Composition
BRONZE 82% Economizers SILVER 67% Perishable Lovers    27% Economizers GOLD 67% Absolute Spenders    30% Perishable Lovers
Python Pandas XGBoost Random Forest K-Means Seaborn
8,504 Observations After 5.5% filter from 9,000
94% XGBoost Accuracy Majority class recall: 99%
15% Optimal Depth 997 of 6,649 customers targeted
178% Projected ROI €5,329 profit · €8,320 total value
03 / Descriptive Analysis

Public Sector UX
& Trust Modelling

Factor analysis and K-Means clustering of 4,239 national survey responses across 42 government services, revealing the “Digital Hater” and “Public Lovers” personas.

Key Cluster Personas
DIGITAL HATER High income & education Low digital service trust PUBLIC LOVERS High public satisfaction ↓ vs. private equivalents VARIANCE EXPLAINED BY FACTOR ANALYSIS 57%
SAS Factor Analysis K-Means Survey Analytics
4,239 Survey Responses National survey dataset
42 Services Evaluated Government digital services
57% Variance Explained Factor analysis output
6 K-Means Clusters Demographic & psychographic segments
04 / Predictive Analytics in Marketing

UTAUT2 Adoption
PLS-SEM Model

Structural equation modelling of technology adoption behaviour using UTAUT2 constructs. Hedonic Motivations identified as the primary driver of Intention to Recommend.

Path Diagram — UTAUT2
PERFORMANCE EXP. EFFORT EXPECTANCY HEDONIC MOTIV. SOCIAL INFLUENCE HABIT BEHAVIORAL INTENTIONS β=0.503 INTENT. RECOM. R²=0.684 R²=0.638
SmartPLS PLS-SEM UTAUT2 Survey Analytics
255 Valid Responses From 381 total · 50/50 gender split
0.684 R² Behav. Intents Q²=0.650 predictive relevance
0.638 R² Intent. Recom. Q²=0.603 predictive relevance
8.782 T-Value (BI→ITR) β=0.503 · Hedonic β=0.350, T=6.317
05 / Market Research

Olive Oil Industry
Market Research

Factor analysis, hierarchical clustering, and correspondence analysis of the Portuguese olive oil market. Portugal ranked 5th globally in 2022 exports at 920M€.

Scree Plot — Eigenvalues
Kaiser F1 3.60 F2 1.37 F3 1.01 F4 0.82 3.6 1.0
SAS Factor Analysis Cluster Analysis Correspondence Analysis
102 Valid Responses Cleaned from 169 total entries
0.7949 KMO Adequacy 3 factors via Kaiser criterion
920M€ Portugal Exports 2022 · 5th globally · 121.9M kg produced
4 Linkage Methods Centroid · Median · Single · Complete
06 / Business Intelligence with Viz Power BI · Dash

Killer Glute Bikes
Executive BI Dashboard

A full-featured executive dashboard combining Power BI and Plotly Dash: profit-by-product matrix with RAG KPI indicators, sales trend bar chart, regional customer data table, and four interactive dropdown filters.

KPI Threshold Indicators
TOTAL ORDER QUANTITY 250K 0 150K Below target Warning zone Target met GROSS PROFIT MARGIN 10.5% 0% 5% NET PROFIT MARGIN 0% 25% 35%
Power BI DAX Python Plotly Dash Pandas
250K Order Qty Target Yellow: 150K–250K · Red: <150K
10.5% Gross Profit Target Yellow: 5–10.5% · Red: <5%
35% Net Profit Target Yellow: 25–35% · Red: <25%
4 Interactive Filters Year · Region · Country · Product Category
3 Dashboard Views Profit matrix · Sales bar · Regional table
RAG KPI Indicators Red / Amber / Green per threshold
02 Data Methodology
01
Sources
CSV files, relational databases, APIs and event streams.
02
Ingestion
Load and catalogue raw source data with provenance tracking.
03
Validation
Enforce schema contracts, detect nulls and referential anomalies.
04
Transformation
SQL-driven cleaning, joining, reshaping and business logic application.
05
Data Model
Dimensional modelling: star schemas, fact and dimension tables.
06
Analytics / ML
Business intelligence, statistical analysis and machine learning.
07
Visualization
Power BI dashboards, DAX measures, interactive reports.
08
Decision
Actionable findings communicated to stakeholders.
03 Technical Arsenal
04 About

I work across data analytics, business intelligence and data engineering, with a focus on turning complex datasets into systems that people can actually use.

Background
Data Analytics
Business Intelligence
Machine Learning
Technical Art / 3D
Beyond Data
Technical Art 3D Modelling VFX Digital Scenography

A background in technical art and visual production shaped my approach to technical problem-solving, visual communication and systems thinking.

Positioning
Data Analytics
Business Intelligence · Power BI · DAX · Analytical storytelling
Data Engineering
SQL · PostgreSQL · Data modelling · Pipelines · Docker
Applied ML
Python · Segmentation · Classification · Sequence mining
Automation
n8n · APIs · Event-driven workflows · Self-hosted infrastructure

Let’s build
something useful.

Download CV GitHub LinkedIn Email