MS Data Science · AWS Certified Data Engineer. Two-plus years building production data systems for Mondelez, Nestlé, L'Oréal and Asahi, from raw ingestion to the decisions they drive.
SQL pipelines. Python automation. LLM-powered agents. Power BI dashboards. I build the full stack from raw data to business decisions. Visualisation is the garnish, the work is upstream.
Owned the full lifecycle of reporting systems, from schema design through deployment. 15,000+ daily records, SQL/Python pipelines and Power BI dashboards used by operations, finance and C-suite for enterprise clients (~$2M Mondelez, Nestlé, L'Oréal).
Enterprise data migration to AWS. SQL extraction & validation, dataset reconciliation at scale. Built the validation workflows that caught errors before they reached downstream systems.
Partnered with operations, finance and C-suite to translate requirements into data products that get used.
SQL + Python pipelines with automated validation gates.
DAX KPI measures across financial, logistics & client performance.
Replaced Excel workflows with Python + SQL + Power Automate.
Anomaly detection, drift monitoring & systematic reconciliation.
Playwright scrapes Seek listings, GPT tailors my resume & cover letter per job, then a live dashboard tracks every application. Background schedulers monitor new listings continuously, with no manual applications. FastAPI + MongoDB backend, Next.js frontend, AWS Cognito auth.
AI inbox analysis over WhatsApp/SMS. Ask in natural language. Claude uses tool-use reasoning to search Gmail, summarise threads, and surface what matters.
End-to-end reporting for Mondelez, Nestlé, L'Oréal. 15,000+ daily records, automated Power BI dashboards across the business.
CNN-based audio classification detecting emotions from speech patterns, with a real-time inference pipeline.
A dashboard is only as trustworthy as the pipeline behind it. These are the rules the upstream work runs on.
Pipelines re-run safely. Same input, same output. No duplicate records, no manual cleanup after a failed job.
Quality gates sit between every stage. Bad data fails loudly and early, before it ever reaches a dashboard or a decision.
Drift monitoring, anomaly detection and systematic reconciliation run continuously, not bolted on after something breaks.
Every manual Excel step is a bug waiting to happen. SQL, Python and Power Automate replace the copy-paste entirely.
Every pipeline I ship exists to shorten the distance between a number and an action. Visualisation is the garnish. The work is upstream.
Looking for roles in data engineering, AI systems and analytics. Melbourne-based, open to relocation, full Australian work rights.