Private AI tools · Internal apps
Private AI for real work.
DropShock Digital · Portfolio · LinkedIn
I build private AI tools and internal apps that make messy team work easier to run.
I turn scattered tools, files, and conversations into systems people can use, understand, and improve.
- Private AI that fits into real work
- Internal tools, shared knowledge, and workflows
- Websites and delivery tools that make services easier to run
I keep a private Linux environment for the work itself: Olympus, the server foundation; Argos, the desktop workspace; and Zeus, the recovery layer.
It gives me a place to test local models, internal apps, automation, media workflows, and self-hosted tools with real constraints. The details stay private; the lessons and public work do not.
- Agent Thread Naming — makes agent work easier to follow.
- Supermemory Local Docker — a community-maintained Docker Compose setup for Supermemory Local.
- Meeting Scribe — self-hosted, consent-aware meeting records for Discord communities.
- LMTokenCook — prepares large folders of text and code for AI tools.
- PhotoPackager — organized, client-ready photo delivery packages.
I participate in communities including OpenClaw, Hermes Agent, AgentWasp, and OpenCoven. I use those spaces to exchange notes, test ideas, and document useful observations; I do not claim official affiliation with—or speak for—those communities.