Archive

Everything, Searchable

All projects, writing, experiments, and milestones in one place.

Use the archive as a fast navigation layer across the merged site. It surfaces retained projects, writing, lab experiments, and timeline entries in one searchable index.

01 · Search & Filter
41 items
02 · Archive Index
projects

ASV Navigation Stack

Planning-Stage Autonomy Architecture for Surface Vessels

2026
projects

Computer Vision Platform

YOLO + FastAPI + React Operator Runtime

2026
projects

3D Printer Guard

Manufacturing Operations Web Platform

2026
projects

GY Corporate Website

Industrial Digital Presence & Content Architecture

2026
projects

Internal Management System

Role-Based Workflow & Records Platform

2026
writing

Connection Lifecycles in Real-Time Vision Systems

Why a streaming endpoint that “works” can still break shutdown, inflate latency, and make device ownership messy.

2026
writing

Memory Is Not Chat History

Why saving conversations is necessary, but still not the same thing as building memory.

2026
writing

Routing Is Not Prompting

Why a working LLM endpoint is not enough once an Assistant shares runtime state, canonical knowledge, and operator workflows.

2026
timeline

Joined SeaPower as an Autonomous Systems Engineer, building autonomous platforms.

2026
projects

Wildfire Surveillance Autonomy

From Fast-Mixing Coverage to Smooth Risk-Aware Aerial Search

2025
projects

ML Competition Portfolio

Selected Medal Cases, Retrieval Work, and Public Artifacts

2025
writing

Quadrotor Surveillance: Local Peak Seeking and Smooth Coverage Control

An early note on unknown utility maps, hybrid control, and why smooth reference generation became the real problem.

2025
timeline

Published key doctoral research outputs on gradient descent dynamics, stochastic coverage, and hybrid optimization for drone surveillance.

2025
writing

Dynamics of Optimization: From Faster Iterations to Faster Settling

An early note on continuous optimization, circuit equivalents, and why settling time became the real design problem.

2024
writing

Exoplanet Signal Extraction: From Raw Sensor Cubes to Smoothed Spectra

How a calibration-first pipeline and moving-average smoothing stabilized a silver-medal Kaggle solution.

2024
writing

Lumbar Spine Classification: From One Global Model to Split Specialists

How decomposing heterogeneous spine targets improved a bronze-medal RSNA competition solution.

2024
timeline

Worked as a part-time Data Scientist at Guangyuan Keyou and reached Kaggle Expert tier.

2024
writing

Markov Matrix Design: From Global SDP to Local ADMM

How a decomposition-first view makes large stochastic-coverage optimization practical.

2023
projects

Dynamic PID Control for Drone Trajectory Optimization

MATLAB Sandbox for Hybrid Optimization, PID Control, and GMM Target Fields

2022
writing

Designing the Markov-Matrix: Why Adapting a Random Walk Fell Short

An early note on Markov chains, swarm density, and why mixing speed became the real problem.

2021
timeline

Began full-time PhD research under competitive FCT funding at Instituto Superior Técnico for wildfire surveillance in drone swarms.

2021
projects

SIR-Model

Control-Oriented Infection Networks Through a Linear-Systems Lens

2020
writing

Linear Systems Thinking for Infection Networks

How a control-oriented model changes how you think about diffusion, sensing, topology discovery, and recovery.

2019
timeline

Started the Ph.D. in Electrical and Computer Engineering at Instituto Superior Técnico.

2019
timeline

Awarded Scholarship Within the Framework of R&D Projects and Institutions (BL204/2018).

2019
timeline

Graduated in Electrical and Computer Engineering from the University of Macau.

2018
timeline

Received the Henry Fok Foundation Scholarship as the top student in ECE.

2015
timeline

Began B.Sc. in Electrical and Computer Engineering at University of Macau.

2014
experiments

Fast-Mixing Policies for Stochastic Coverage

A lab thread on designing Markov guidance laws that do more than match a target ...

experiments

Local ADMM for Markov-Matrix Design

A decomposition-first lab note on turning one large stochastic-coverage SDP into...

experiments

Settling-Time Design for Accelerated Optimization Flows

A continuous-time lab thread on reinterpreting accelerated optimization as dynam...

experiments

Circuit Equivalents of Optimization Dynamics

Exploring when optimization laws can be expressed as circuit-like continuous sys...

experiments

Smooth Coverage over Unknown Utility Fields

A local-surveillance lab thread on turning only local gradient information into ...

experiments

Restart Logic for Multi-Peak Aerial Search

A hybrid-surveillance experiment on why local ascent alone is not enough in stro...

experiments

Formation-Aware Distributed Surveillance

An ongoing extension of the wildfire research line from area redistribution towa...

experiments

Source Localization from Partial Infection Observations

A compact lab study on how much one can recover about an infection source when o...

experiments

Topology Recovery in Linear Infection Networks

A modeling thread on how the linear-systems view makes topology discovery, local...

experiments

Spectral Denoising for Exoplanet Retrieval

A competition lab note on why a calibration-first, smoothing-heavy pipeline outp...

experiments

Split Specialists for Lumbar MRI Classification

A competition lab note on decomposing a heterogeneous medical-imaging task into ...

experiments

Agentic Assistant for a CV Operator Platform

An active systems investigation into how an LLM assistant should sit inside a Fa...

experiments

Disconnect-Aware Streaming in a CV Runtime

A platform investigation on why shutdown latency, camera ownership, and seemingl...