I connect data, models, APIs, and interfaces into useful products, from prediction tools to support workspaces and production-ready web apps.
My name is Fernando Lin. I build at the intersection of software engineering, data, and AI, with a focus on turning technical systems into interfaces people can actually use. I like exploring datasets, testing new frameworks, and shaping ideas into functional web products.
Passionate about creating efficient and innovative web solutions using Next.js, Django and Laravel.
Skilled in exploring and analyzing datasets to uncover valuable insights using Pandas, NumPy, and visualization tools.
Focused on machine learning, natural language processing, and computer vision using TensorFlow and Scikit-learn.
The tools I use are organized by how I turn an idea into a working product.
My strongest overlap is full-stack product engineering with AI and analytics woven into the workflow, especially tools that need a clean interface around a model, dataset, or knowledge base.
Each case shows what changed: before, what I built, and the impact after.
Risk prediction lived as raw model output without an approachable workflow for users.
Flask app, neural network inference, SHAP explainability, and tabular data processing.
Patient inputs became readable risk signals with SHAP context for easier interpretation.
This project is a web-based application designed to predict the risk of heart disease using a neural network model.
Project work was scattered across repos without a clear story or hiring-focused path.
Next.js interface with motion, responsive sections, and curated project data.
Visitors now get one focused path through my AI, data, and full-stack work.
A personal portfolio website that showcases my skills and projects.
Rental details were hard to browse without a structured vehicle and pricing experience.
Laravel, Blade templates, SCSS styling, and JavaScript interactions.
Customers can review vehicles, compare pricing, and move toward booking in one flow.
A car rental website that provides information about cars, courses, pricing, and booking options.
The food brand needed a clearer way to present products beyond static listings.
Next.js, React, TailwindCSS, and polished motion for menu and product discovery.
The brand gained a deployed product showcase with faster menu and product discovery.
Restaurant website that sells Izakaya Suki products.
Support operators had to jump between product knowledge, policy, and customer context.
LLM workflow, RAG-style product knowledge, analytics surfaces, and a Next.js workspace.
Operators can answer faster from one AI workspace connected to knowledge and policy.
CircuitDesk combines AI assistance, structured product knowledge, customer support policy, and operator analytics in one polished workspace.
Manual calorie tracking forced users to search foods, estimate portions, and type every meal.
Next.js, React, TailwindCSS, computer vision, and Supabase-backed meal records.
Photo recognition reduced logging time by 80% by turning meal pictures into calorie insights.
An AI calorie tracker that estimates meal nutrition from food photos.
A live contribution signal from my GitHub profile.
Send a project idea, collaboration note, or question into my inbox.