HE FENGYU

University of Tsukuba · Multimodal AI · iOS Engineering

HE FENGYU

Researcher / iOS Engineer / Master Student

Building thoughtful systems at the intersection of multimodal AI, explainability, and human-centered software.

View Projects Read Research Contact Me

About

Research taste, engineering discipline, and a quiet eye for detail.

I am a master student in Library and Information Science at the University of Tsukuba, working around multimodal AI, explainability, and SNS impression analysis. My engineering practice spans iOS, backend systems, cloud deployment, and AI-assisted research tools.

Research

Explaining how people read meaning, trust, and intent in social posts.

01

Multimodal Impression Analysis for SNS Posts

Modeling how text, imagery, layout, and platform cues shape perceived advertisement-like quality, trust, experience, and closeness.

Text Image SNS
02

Explainable AI for Advertisement-like Perception

Designing interpretable models that surface the visual and linguistic cues behind impression judgments instead of returning opaque scores.

XAI Trust Cues
03

Text-Image-OCR Fusion and Concept/Cue Bottleneck Models

Exploring fusion pipelines where OCR, captions, concepts, and cues become structured intermediates for more understandable predictions.

OCR Fusion Bottleneck

Projects

Product-minded engineering across mobile, backend, cloud, and AI tools.

Apexing app mark

Apexing / 弯心

Active

A driving record and route challenge app for capturing speed, heart rate, sensor data, and route experiences.

Swift UIKit SwiftUI Go Cloud Run HealthKit WatchKit
Open project

Personal Website

Live

A modern portfolio for research, engineering projects, photography, and writing.

HTML CSS JavaScript Responsive UI
View current site

AI Annotation Tools

Research

Tools for multimodal data annotation and research workflow automation.

Python LLM Web Data Processing
Ask for details

Experience

A timeline shaped by software, research, and human-centered systems.

  1. Master student, University of Tsukuba

    Researching multimodal impression analysis, explainable models, and SNS perception.

  2. Apexing product development

    Building iOS, backend, and cloud systems for driving records, route challenges, and sensor-driven experiences.

  3. iOS Engineer experience

    Product engineering background with a focus on polished mobile interaction and maintainable implementation.

  4. Software engineering background

    Practical stack across Swift, Go, Python, web development, and cloud deployment.

Photography / Life

Notes from roads, cities, cameras, and ordinary light.

Evening Desk Research, camera, and late-day light.
Route Memory Driving data and road impressions.
City Study Urban texture, signage, and small scenes.
Camera Notes A replaceable slot for future photographs.

Contact

Research notes, product ideas, and thoughtful collaborations are welcome.

The fastest way to reach me is email. You can also find writing, code, and photography through the links below.

GitHub Blog X / Twitter LinkedIn
Email copied