Clarence
  • Home
  • About
  • Projects
  • Featured Posts
  • Experiences
  • More
    Publications Skills Education
  • Posts
  • Notes
  • Blog
  • English
    繁體中文 English 日本語
  • Dark Theme
    Light Theme Dark Theme System Theme
  • Tags
    • AI
    • Automation
    • Backend
    • Backup
    • CI/CD
    • Cloudflare
    • CUDA
    • DevOps
    • Docker
    • GitLab
    • GPU
    • Harbor
    • Hosting
    • Infrastructure
    • Kaldi
    • LLM
    • Monitoring
    • Network
    • Nginx
    • Ollama
    • On-Prem
    • OpenAI
    • Operations
    • Proxmox
    • Python
    • Speech AI
    • Whisper
2026-10-26

Designing an LLM Summarization Pipeline: OpenAI API, Ollama, and Local Models

The principles and trade-offs I use when designing and operating an LLM text summarization pipeline, covering preprocessing and chunking, cloud API versus local models, batch processing, retries, cost control, and output validation.

  • LLM
  • Ollama
  • OpenAI
  • Python
  • AI
Read →
2026-07-08

How I Work: Backend, AI, DevOps, and Infrastructure

How I connect Backend, LLM / Speech AI, DevOps, Cloud, On-prem Infrastructure, and Network Operations into one delivery chain.

  • Backend
  • LLM
  • Speech AI
  • DevOps
  • Infrastructure
  • Network
Read →
Navigation
  • About
  • Projects
  • Featured Posts
  • Experiences
  • Publications
  • Skills
  • Education
  • Blog
Contact me:
  • me@clarence.tw
  • clarencetw
  • Clarence
  • Clarence

Toha Theme Logo Toha
© 2026 Copyright.
Powered by Hugo Logo