Debby's

Featured project:

How I work. To me, a project isn’t done when the code runs. It’s finished someone makes a better decision because of it. I love owning the whole journey. From architecting the backend to pitching the final demo in the boardroom, I bridge the gap between the terminal and the stakeholders to ensure the tech doesn’t just run, but actually lands with the users. Not because I can’t specialize, but because I believe the best systems are built by people who understand every layer.

I build the plumbing

Turning messy data swamps into trusted, single-source-of-truth systems. I’ve stood up AWS environments, automated ETL pipelines that replaced 2 hours of manual Excel pain with a 5-minute script, and maintained the certified data sources behind dashboards serving 1,300+ users.

I make AI earn its keep

Problem first, technology second. Unlike traditional software, AI doesn’t scale for free. Every API call costs money. Every token adds up. So I think hard about cost structure from day one. I designed a Text2SQL engine that lets non-technical colleagues query databases in plain English. Not because LLMs are cool, but because the alternative was a 3-day turnaround from an analyst for every questions.

I translate for both the Boardroom and the terminal

A decade of consulting taught me that the best technical work is useless if nobody understands it. I’ve presented to 1,000 delegates at an international congress, trained teams on AWS and SQL, and built dashboards that drove 138% higher engagement than platform average. I’m equally comfortable in the terminal and the boardroom and I know which room needs which language

Translating technical intelligence
into strategic business value

Data & Machine Learning

Built hybrid BERT/GPT-4 deployment architecture achieving GPT-4 accuracy at 16% less cost/li>
Engineered ARIMA forecasting system for housing analysis (1-2% MAPE vs industry standard 5-12%)
Designed NLP pipeline processing 30+ years of policy documents using Python and OpenAI API

Architecting the engines
of automated innovation

Tech & Architecture

Architected a security-first AWS cloud data environment with automated data quality frameworks that was built to enterprise compliance standards from day one
Engineered ETL/ELT pipelines using PySpark and SQL, reducing processing time from 2 hours to 5 minutes (96% improvement)
Designed Text2SQL2Chart pipeline using LLM for natural language data access

I believe the best way to learn is to do. “A mix of production systems, competition winners, and weekend experiments. Some projects are anonymized for confidentiality but the engineering is real.

SmartSaver: Bank Calculator Tool

Data Tech

Show project

Birdhouse: Data-driven Senior Care Monitoring

Development

Show project

Topic Modelling of Song Lyrics to Predict Billboard Longevity

Data Tech

Show project

From Rail to Real Estate: Award winning Tableau Dashboard

Data Tech

Show project

Young ESOMAR Society Award

Data Tech

Show project

SenseCheck: AI-Powered Comms Stress-Testing

Data Tech

Show project

part builder
part data story-teller
full-time problem-solver

Hello,
I'm Debby

SmartSaver
Interest Calculation App

I spent years as a consultant telling stories with data...

...now, I engineer the
systems that make those
stories possible

I build the plumbing

I make AI earn its keep

I translate for both the Boardroom and the terminal