Think of this portfolio as three chapters:
Agentic AI Products
Designing and deploying agentic AI companions used by real people: across cooking, career development, and newcomer integration.
SofIA - Your Ally in Canada
Phase IA bilingual AI companion for Latino newcomers navigating the Canadian job market. SofIA helps users understand why reaching out to other professionals matters, and actually enjoy doing it.
Built with real user research across 9+ newcomer interviews, it delivers personalized goal-setting, regional resources, and culturally-aware guidance in English and Spanish.
Developed as the project for the Aggregate Intellect Agentic AI Buildcamp.
Tools: Python, Claude (Anthropic), FastAPI, Twilio (WhatsApp), Firestore, GCP Cloud Run, Docker, Langfuse
Personal CookFlow AI Agent
This project was developed as part of the Aggregate Intellect Agentic AI Bootcamp, which focused on building
production-ready multi-agent AI systems grounded in real user research.
CookFlow is an AI-powered kitchen assistant that plans meals around what you already have, eliminates food waste, and
delivers a personalized cooking plan, all from a single conversation.
Tools: Gemini API, Google ADK, Python, FastAPI,GCP Cloud Run, Firestore
Data Science & Consulting
I have worked on some exciting projects, including consulting, challenges and my Ph.D. thesis
Ph.D. Thesis
My PhD research at Western University is part of a collaborative project with a major Canadian Oil & Gas company. The objective of the project was to apply mathematical modelling and ML techniques to better understand hydrocarbon/water mixtures.
Tools: Python, Pandas, Scikit-learn, Mathematical modelling, Statistical analysis
Recruit WebApp
I was the Lead Developer of an API to help a local school bus company in London, ON. The WebApp helped to improve and keep accountable the recruitment and training processes of the organization.
Tools: Flask (Python), SQL + SQLAlchemy, Pandas, HTML, CSS, JQuery, Pythonanywhere
Drain Tracker
I am the Lead Developer for an asset management tool for a small municipality near London. We are developing a WebApp to help the superintendent of drains manage the repairs and builing of the drains infrastructure. This project will involve migration of 20 years of data from the old system.
Competitions & Playground
During my free time I enjoy upskilling and putting in practice my skills by playing with data and taking online courses.
AI Multi-Agent: Customer Segmentation & Marketing Planner
This project was created as part of the Google DevFest 2025 London: Building Safe, Secure and Scalable Solutions with AI and Cloud event.
I directed a 90 minutes interactive workshop: Agentic AI: Hands-On with Gemini in Kaggle
Tools: Gemini API, Google ADK, Python
AI-Powered Financial Report Summarizer
This project was developed as the capstone for the 5-Day Gen AI Intensive Course with Google.
Financial reports are dense and time-consuming to analyze. This project uses Retrieval-Augmented Generation (RAG) with Gemini to automatically generate concise, context-aware summaries by combining document retrieval with multimodal reasoning. The output includes a structured summary table for quick insight.
Tools: Gemini API, Python
Smartsito
Our team, “SmartSito” was one of the top 3 finalists in the Retail & E-commerce category for the 2020 Microsoft Discover AI Challenge: Smarter and Sustainable Economies.
Smartsito attempts to reduce food waste using computer vision and a reward system.
Tools: Flask (Python),Pandas, Azure Cognitive Services, Azure Websites
AI Upskilling Journey
Our team "Team Lopez AI" was the winner for the 6th Sprint of Microsoft AI Upskilling Journey.
Tools: Flask (Python), Microsoft Azure: Cloud Computing Services, Azure Cognitive Services, Azure Websites
Logeo
Our team "Logeo" developed a webapp that transform 2D logos to 3D using RigNet model.
I was responsible for the deployment of the REST API.
Tools: Flask (Python), AWS EC2 instance, RigNet
Topic Modelling of Banking Transactions
In this notebook, I analyze bank statements from my everyday banking account and my two credit cards using Topic Modelling. The idea is to classiffy the transactions. I also take a look on how my transactions have been affected by COVID 19.
Tools: Pandas, Scikit-Learn (LDA), Seaborn, Plotly
Customer Segmentation
In this notebook I perform customer segmentation using a credit card dataset from Kaggle. K-Means is used for the clustering.
Tools: Pandas, Scikit-Learn (KMeans), Seaborn