SeisBlue Co., Ltd. logo SeisBlue SeisBlue Co., Ltd.

SeisBlue Co., Ltd. · Taiwan

Everything starts
with software & AI.

SeisBlue Co., Ltd. is a Taiwan-based technology company built around software engineering and AI. From that core, we extend the same capability into seismic AI applications and brand web design & marketing.

Explore our work Contact us →

About SeisBlue

Software & AI at the core.
Everything else builds on it.

SeisBlue's core capability is software engineering and AI. We bring this capability into different domains to solve real-world problems.

We're always doing the same thing: turning abstract requirements into deliverable systems.

What We Do

One core capability,
with its extended applications.

SeisBlue is grounded in software and AI engineering. From this core, the same capability extends to different contexts and clients.

Technical Core

Jimmy Lab
Software & AI Consulting

SeisBlue's engineering core. AI development, Vibe coding workflows, Cloudflare Tunnel / Tailscale infrastructure, and automation projects — the public-facing technical blog and consulting window.

Visit jimmylab.seisblue.com ↗
AI Application

SeisBlue
Seismic AI

Applying our AI engineering capability to seismic monitoring. Deep-learning phase picking, near-real-time earthquake catalogs, and early warning systems.

See the technical details →
Web & Marketing

Blue's Reserve
Brand Portfolio

A custom beverage brand we built and operate ourselves — from the product site and visual design to the marketing workflow. A complete showcase of our web and brand marketing capability.

Visit reserve.seisblue.com ↗

AI Application · Seismic AI

From original methods,
to production deployment.

In seismic AI, SeisBlue covers the full stack — from methodology research to production deployment. We design original training methods, build training pipelines, and turn research models into services that run reliably in production. Below are concrete demonstrations.

01

Original Methodology

New loss functions, training strategies, and data engineering — raising performance ceilings at the model level.

02

Training Infrastructure

Reproducible, scalable training infrastructure — covering data preprocessing, model architecture, and training strategy.

03

Production Deployment

Integrating models into real-time data streams (e.g. Earthworm) to keep services running 24/7.

Case 1 · BlueDisc: Original Methodology Research

BlueDisc is an AI core methodology research project conceived and led entirely by SeisBlue. We address the long-standing S-wave amplitude suppression problem in deep-learning phase pickers — from problem diagnosis to a shape-then-align training strategy, validated through a conditional GAN proof of concept. On standard benchmarks, effective S-phase detections rose by 64%. Full results in the preprint below.

Case 2 · The SeisBlue Platform

SeisBlue is a deep-learning seismic data processing framework led and developed by us. The AI core architecture and model design are provided by our company. The platform packages the full pipeline — from waveform ingestion through phase picking to catalog output — into a service that runs reliably in production. It has been adopted by multiple peer-reviewed studies as a core processing tool (see citations below).

Case 3 · TT-SAM Real-Time EEW Pipeline

TT-SAM (Taiwan Transformer Shaking Alert Model) is an earthquake early warning system. Its architecture references the TEAM framework by Münchmeyer et al. (2020); SeisBlue led the AI engineering core — training environment setup, pipeline design and guidance, and production deployment. The resulting TTSAM Realtime pipeline has been delivered to Taiwan's Central Weather Administration (CWA), running stably on the Earthworm real-time data stream.

Academic References

Below are the primary publications related to SeisBlue and TT-SAM, along with follow-up research adopting these technologies.

BlueDisc · Original Methodology

  1. Lead Technical

    Huang, C.-M., Chang, L.-H., Chang, I.-H., Lee, A.-S., & Kuo-Chen, H. (2025). Recovering Sub-threshold S-wave Arrivals in Deep Learning Phase Pickers via Shape-Aware Loss. arXiv preprint, arXiv:2511.06731. arxiv.org/abs/2511.06731

SeisBlue Platform

  1. Primary

    Huang, C.-M., Chang, L.-H., Kuo-Chen, H., & Zhuang, Y. (2023). SeisBlue: a deep-learning data processing platform for seismology. EGU General Assembly 2023, Vienna, Austria, EGU23-13927. doi.org/10.5194/egusphere-egu23-13927

TT-SAM System

  1. Primary

    Chen, et al. (2026). A Deep Learning Framework for Peak Ground Velocity Prediction Using Multi-Station Velocity Waveforms: The Taiwan Transformer Shaking Alert Model (TT-SAM). Journal of Geophysical Research: Machine Learning and Computation. doi.org/10.1029/2025JH001005

  2. Architectural Reference

    Münchmeyer, J., Bindi, D., Leser, U., & Tilmann, F. (2021). The transformer earthquake alerting model: A new versatile approach to earthquake early warning. Geophysical Journal International, 225(1), 646–656. doi.org/10.1093/gji/ggaa609

Follow-up research adopting the SeisBlue Platform

  1. Kuo-Chen, H., et al. (2024). Deep learning-based earthquake catalog reveals the seismogenic structures of the 2022 MW 6.9 Chihshang earthquake sequence. Terrestrial, Atmospheric and Oceanic Sciences. doi.org/10.1007/s44195-024-00063-9

  2. (2023). The first 30 min hidden aftershocks of the 2022 September 17, ML 6.4, Guanshan, Taiwan earthquake and its seismological implications. Terrestrial, Atmospheric and Oceanic Sciences. doi.org/10.1007/s44195-023-00059-x

  3. Sun, W.-F., Pan, S.-Y., Liu, Y.-H., Kuo-Chen, H., Ku, C.-S., Lin, C.-M., & Fu, C.-C. (2025). A Deep-Learning-Based Real-Time Microearthquake Monitoring System (RT-MEMS) for Taiwan. Sensors, 25(11), 3353. doi.org/10.3390/s25113353

  4. (2025). Real-Time Earthquake Monitoring with Deep Learning: A Case Study of the 2025 ML 6.4 Dapu Earthquake and Its Fault System in Southwestern Taiwan. The Seismic Record, 5(3), 320. geoscienceworld.org

Technical Core · Jimmy Lab

SeisBlue's engineering core,
our consulting window.

Jimmy Lab is the public showcase of SeisBlue's technical core, and our window for taking on AI development, automation, and infrastructure projects. The software and AI capabilities behind every external application are accumulated here.

Visit jimmylab.seisblue.com ↗

Main Topics

  • AI Development WorkflowsVibe coding, Codex, Claude Code, agent team management
  • Infrastructure & TunnelingCloudflare Tunnel, Tailscale VPN, self-hosted services
  • Automation & ToolingDev environment setup, CI/CD, performance optimization
  • ConsultingAI development, backend, automation, technical advisory

Web & Marketing · Blue's Reserve

Web design and brand marketing,
packaged in a bottle.

Blue's Reserve is a custom beverage brand built and operated entirely by SeisBlue — and a complete showcase of our web design, product page, and brand marketing workflow. From drink selection, label design, and bottle pairing to the product site and LINE inquiry flow, the work is publicly accessible at reserve.seisblue.com.

Visit reserve.seisblue.com ↗

What We Offer

  • Custom DrinksBeer, sparkling honey wine, fruit wines, non-alcoholic
  • Bottles & ContainersAluminum cans, glass bottles, beer bottles
  • Label DesignVisual design and small-batch printing with 3D preview
  • Use CasesCorporate gifts, brand pop-ups, local rites, curated retail

Contact

Work with SeisBlue.
Start with an email.

Whether you're looking at seismic monitoring collaborations, software development projects, or a custom drink for a brand event, just send us a message.

SeisBlue Co., Ltd. · 藍震有限公司