Welcome to UpWave Research

Ocean waves emerge at the boundary of atmosphere and sea — complex, stochastic, yet governed by deep physical laws. At UpWave Research, we apply the same modeling instinct to physical systems, financial markets, and scientific communities alike.

Discover More

What We Do

AI-native research across physical systems, financial markets, and scientific communities

🔬

Numerical Modeling

Building quantitative models for physical systems — atmosphere, ocean, climate — and financial markets. From discretized fluid dynamics to stochastic processes, we turn complexity into tractable computation.

🧠

Data Intelligence

Extracting high-value signals from massive datasets. AI-driven pattern recognition and decision support across scientific and financial domains.

⚙️

Scientific Tool Automation

One-stop automated solutions that make cutting-edge research tools accessible. See our work on pyCFRAM — streamlining climate feedback analysis for the research community.

How We Work

Lean research, AI-amplified

A focused, lean team — amplified by a constellation of AI agents. Research tasks are orchestrated in parallel, analysis automated end-to-end. This operating model lets us deliver research-grade output across multiple domains without sacrificing depth or rigor.

Global Presence

Distributed infrastructure supporting global research and market operations

🇭🇰

Hong Kong HQ

Our headquarters is located in Hong Kong, Asia's premier international financial center.

🌐

Distributed Servers

Server infrastructure strategically deployed across United States, Singapore, and Japan for optimal performance.

⚡

Real-time Processing

Low-latency data processing and execution across multiple time zones and markets.

Featured Projects

MercuriusLite

A Lightweight Quantitative Investment System

A self-built quantitative system focused on asset allocation of macro asset classes in personal investment portfolios. Provides objective guidance for strategy developers.

Learn More →
🌍

cmip6-to-wrfinterm

CMIP6 to WRF Regional Downscaling Pipeline

Automated one-stop conversion of CMIP6 global climate model output into WRF-compatible intermediate files for regional dynamical downscaling. Handles multi-layer soil data, non-standard calendars, and multiple GCMs including MPI-ESM, EC-Earth3, and CESM2.

View on GitHub →
🌡️

pyCFRAM

Climate Feedback-Response Analysis Method

Python implementation of CFRAM, decomposing climate model temperature changes into individual physical drivers — CO₂, water vapor, clouds, aerosols, surface albedo, and more. Dual radiation engines (RRTMG and Fu schemes) with multiprocessing acceleration for CMIP6-scale analysis.

View on GitHub →