Leading product architecture and frontend development for an AI-powered lung auscultation platform helping physicians detect respiratory disease faster. Led a team across data collection and product implementation, supporting the clinical research pipeline and translating findings into scalable product features.
I’m Marvin Patrick Lacanilao
Software Engineer | UI/UX Designer | Frontend Developer
I build digital experiences from idea to product. I work across software engineering, product development, and UX, turning problems into useful, thoughtful solutions. Whether I’m designing an interface, developing an application, or exploring an AI-driven solution, I care about both how something works and how people experience it.
Experience
Graduated Cum Laude, Dean's List 2022–2026
Validated business rules and field-level logic across multiple ERP modules for multiple clients, spanning preparation, integration testing, quality control, and retesting. Executed structured test scenarios throughout the QA pipeline, documenting findings and verifying bug resolutions while collaborating with analysts, developers, and project teams across multiple client engagements.
Projects
More01 / WEB APPLICATION
Colume
A study app that instantly turns notes and PDFs into quizzes and flashcards, featuring a virtual pet companion that levels up with daily study streaks.
02 / WEB APPLICATION
Antlas
A personal job search tracker built to replace messy spreadsheets. Features a visual Kanban board, bento dashboard, application deadlines, and analytics to track applications from submission to offer.
03 / RESEARCH
Dual-Channel CNN-LSTM with MFCC-Chroma Attention Fusion for Lung Sound Disease Classification
Lung auscultation remains highly subjective, with pulmonologists averaging only 36.5% correct detection rates. This study proposes a dual-channel CNN-LSTM architecture with a constrained MFCC-Chroma gated fusion mechanism for multi-class respiratory sound classification. HINGA V1, the first localized Filipino lung sound dataset, was collected from 51 patients across four Baguio City institutions and combined with the ICBHI 2017 benchmark. The model achieved a macro F1 of 0.4056 across six disease classes on unseen data, with the fusion mechanism consistently favoring MFCC-derived temporal features, supporting a viable path toward a clinically relevant auscultation tool for the Philippine healthcare setting.
04 / DESIGN
SpeechScape Cordillera
An interactive linguistic mapping platform dedicated to documenting, preserving, and exploring the diverse indigenous languages, ethnic groups, and cultural heritage of the Cordillera Administrative Region.
define.develop.design.deliver.define.develop.design.deliver.define.develop.design.deliver.define.develop.design.deliver.