On AI Lounge since Nov 2026
Everald Tah
Member #4,088Tech LeaderL2Veraldlabs, Veraldlabs
Enthusiastic Full stack web deb , currently studying cybersecurity and building saas company from home solo , most importantly having fun
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
No posts yet
Everald Tah hasn't posted yet.
Pinned projects
Claude Code on any local model on this machine - every GGUF across claudelocal, LM Studio and ollama, switchable from /model without restarting. Memory planner for a 6 GB card.
Claude Code on local LM Studio models, sized for your GPU. VRAM-aware MoE loader + auto-tuner: gpt-oss-20b and Qwen3.5-35B-A3B on a 6 GB card.
An Odyssey-class coding harness for Qwen 2.5 7B - local-first CLI coding agent combining Odysseus orchestration + opencode UX, optimized for small models.
Qwenodyssey 2.0: a coding-agent CLI for local and cloud models, rewritten in Rust (one input owner, so the /model freeze can't come back)
Native iOS terminal running Claude Code in-process (nodejs-mobile, virtual TTY, SwiftTerm); Linux layer next
About
Enthusiastic Full stack web deb , currently studying cybersecurity and building saas company from home solo , most importantly having fun
