Assembling Systems, Not Building Semiconductors — A New Learning Path in the AI Era
Entangelk
Command Palette
Search for a command to run...
Archive
22 posts published so far
Korean TTS Workbench — (5) Qwen3-TTS → VoxCPM2, Swapping the Model Out
Entangelk
Korean TTS Workbench — (4) When the Workarounds Failed, a Hardcoded Line Inside the Library
Entangelk
Korean TTS Workbench — (3) The Last Syllable Sounds Cut Off — and the Cause Was the Model Itself
Entangelk
Korean TTS Workbench — (2) Korean-Specific Tricks and Running on 4GB VRAM
Entangelk
Korean TTS Workbench — (1) Starting With Qwen3-TTS on a 4GB GPU
Entangelk
Memory Architecture Still Needed in the LLM Era — (3) Interface, Performance, and Where Planning Meets Engineering
Entangelk
Memory Architecture Still Needed in the LLM Era — (2) Three Problems, Three Implementations
Entangelk
Memory Architecture Still Needed in the LLM Era — (1) Why Memory Isn't Just a Log
Entangelk
Where This Design Could Go — The Harness Platform Possibility
Entangelk
Measuring and Deploying Models — Profile-Based Selection and Matching
Entangelk
Design Done, Now Prove It — IR Lowering POC Implementation Plan
Entangelk
Verifying AI with Game Theory — The Adversarial Assurance Loop
Entangelk
Merging Methodology and Infrastructure — Integrated Design & Execution Strategy
Entangelk
From Philosophy to Infrastructure — Enterprise MAS Design
Entangelk
When My Idea Met AI — Role-IR Design Philosophy
Entangelk
How I Recovered a Broken AI-Delegated Data Pipeline in Two Days
Entangelk
In the Age of AI Depression, What a Lazy INFJ Found Instead of Helplessness
Entangelk
When AI Ignores Its Own Rules — Options Beyond Prompt Engineering
Entangelk
Beyond Coding, the Age of the Conductor Begins
Entangelk
The Sweet Spot of AI Orchestration: From AWS Limits to a Home Server Architecture
Entangelk
The Reality of AI-Driven Coding: Humans Still Debug, Decide, and Verify
Entangelk