AI & Computer Vision · Full-Stack Development · IoT
IT Engineering student at UTMACH · Building at LarvIA · Machala, Ecuador 🇪🇨
- 🔬 Applied research in computer vision: deep-learning individual tree detection in airborne LiDAR (TreeSeg-ALS), reproducibility code for a manuscript submitted to Forests (MDPI).
- 🎤 Speaker at FLISOL 2026 — improving small-object detection in aerial imagery with YOLO + SAHI, evaluated on VisDrone.
- 📡 I like building end to end: SISCA, a BLE access-control system with on-device face recognition, spans ESP32 firmware, native Android & iOS apps, and a FastAPI backend.
- 🌐 More about me at jossuema.tech.
| Project | What it is | Stack |
|---|---|---|
| TreeSeg-ALS | Inventory-anchored evaluation of deep-learning tree detection in airborne LiDAR (ALS), with label-efficient cross-domain transfer in alpine forests | Python · PyTorch · PointNet++ |
| SISCA — Android · iOS · Firmware · API | BLE access control for UTMACH: ESP32 (NimBLE) firmware, native apps with on-device face recognition (ML Kit + TFLite) and biometric auth, REST backend | Kotlin · SwiftUI · C++ · FastAPI |
| AquaLens AI | Desktop app that detects and counts fish in underwater video with a two-phase OpenCV + YOLO pipeline | Python · OpenCV · YOLO |
| YOLO + SAHI | FLISOL 2026 talk & demo: small-object detection for urban aerial imagery with real mAP metrics | Python · YOLO · SAHI |
| Buzón Inteligente UTMACH | University participation platform (complaints, suggestions, proposals) with an admin analytics dashboard | Next.js · Prisma · PostgreSQL |
Languages
AI / Computer Vision
Web
Mobile & IoT
