AI
n8n workflows, Telegram bots, local LLMs, the Mem0 memory stack and TTS — AI automations I have actually shipped, mostly on self-hosted hardware.
11 posts in this category
-
Running AI Video Locally on a 12 GB GPU: Real Numbers for Wan 2.2 and SeedVR2
Measured on an RTX 3060: Wan 2.2 TI2V-5B generation, SeedVR2 upscaling, the 1080p resolution trap, and the one command that fixed five failed runs.
-
Why My AI Songs Kept Losing Their Endings: Running YuE2 on a 12 GB GPU
Twelve Mandarin songs came out exactly 240.000 seconds long with the outros missing. The cause wasn't the lyrics — it was a hard cap, a VRAM ceiling, and a planner that ignores BPM.
-
The Five-Step AI Roadmap Everyone Shares — and What I Actually Run in Production
The five-step AI roadmap circulates on every feed. Here is the honest version: which steps I run in production, which one I skipped, and what each is for.
-
Why I Still Bought a GPU to Run AI When Claude and GPT Are Obviously Stronger
Claude Opus and GPT are stronger than anything I can run locally. Here is the cost, control, and uptime argument for owning a GPU anyway — with real numbers.
-
I Ran a Text-to-Speech Service for a Year on Two Cron Containers
A reading app needed TTS. Cloud free tiers expire tokens in 10 minutes and 1 hour. Here's the sidecar pattern that kept Azure and Google voices working for a year with no secrets in any workflow.
-
Best AI Video Generators in 2026: Free vs Paid, and My LTX-2.3 Test
Compare the best AI video generators in 2026 — free vs paid plans, real pricing, local open-weight models, and my hands-on LTX-2.3 test.
-
Using Chinese LLM APIs From Malaysia: What Actually Works
Using Chinese LLM APIs (GLM-5.3, DeepSeek V4) from Malaysia: how I paid in RMB, handled phone verification, and which providers actually work.
-
Slashing Agent Context and Token Churn With a Code Graph
Why dumping source files into LLM prompts leads to token bankruptcy and hallucinated bugs, and how pre-indexed AST knowledge graphs slash context by 80%.
-
Plugging Windows Desktop Control into Four AI Agents: Hermes, OpenCode, Claw, and Claude
How to configure, verify, and standardize windows-mcp across Hermes Agent, OpenCode CLI, Claw Code, and Claude Code without config drift or tool sprawl.
-
Giving AI Your Mouse and Keyboard: Safe Guardrails for Windows Desktop MCP
Why running desktop automation MCPs out of the box is dangerous, and the essential guardrails needed to prevent system corruption, UI hangs, and token blowups.
-
I Let an Agent Manage My Shopee Vouchers — And It Created One I Never Approved
Vouchers have no draft state: Confirm means live and drawing on escrow. Here is the audit → propose → auto design I use to keep an agent away from that button, and the list-lag mistake that cost me an unplanned voucher.
