-
Simbi Community Dev, Minuscule Ventures
- Sacramento, CA
-
23:03
(UTC -12:00) - akapug.net
- @davidakapug
- in/davidryalanderson
Lists (12)
Sort Name ascending (A-Z)
🧠 Agent Memory & State
Stateful persistent memory patterns for AI agents (Letta, agi-memory, etc)🤖 Agent Orchestration
Agent coordination, messaging, beads, A2A patterns💬 Discord & Chat
Discord bots, chat integrations, kimaki patternsFrameworks
🔮 Future ideas
Homelab
for optimizing local AI🏗️ Infrastructure
Convex, SST, deployment, backend patternsLocal AI
Stars
A desktop GUI for CLIProxyAPI and a tool for automatically configuring popular AI agents.
A desktop VM for agents: wrap's base image + XFCE, shared Chrome, agent-browser and agent-desktop. Containerfile + Smolfile.
NInfer with NVFP4 DSpine for RTX 5090 on Windows / Linux.
Abliterated DeepSeek-V4.1-Flash-1017 (FP8 dense, FP4 experts) on 4x RTX PRO 6000 with vLLM-Moet r21, DSpark + FLy verification and top_p 1.0: 463 tok/s single-stream agent decode at 30k-120k contex…
Debloat any Android or Google TV over Wi-Fi and get an Apple TV-style home.
Check whether your self-hosted LLM performs as expected.
Reproducible Docker/vLLM deployment, tuning, and benchmarks for DeepSeek V4.1 Flash on a switchless three-node or four-node DGX Spark fabric.
Ajay9o9 / llama.cpp
Forked from ggml-org/llama.cppLLM inference in C/C++
A small, fast container runtime for macOS, Windows and Linux with free Ngrok alternative
Inter-session communication extension for pi coding agent
Agent Skill for fine-tuning an agentic model on your own Claude Code sessions with Tinker. Works in Claude Code, Codex and OpenCode.
A brain that learns from live experience. System 1 by default: perception, memory, plasticity and action. Optional System 2 for self-observation and slow corrections.
Skills I use to execute long-running coding agents without breaking my back.
LLM serving engines that SkyDiscover builds on the fly for specific AI models and hardware: Qwen3.6-35B-A3B, Qwen3.8-Flash-Next and Flint on Apple silicon Macs
SpecPilot: Autonomous Algorithm Discovery and Systems Optimization for Speculative Decoding
Qwen3.5-2B with a filesystem compiled and inserted directly into its weights.
Code for Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention
Tiny self-hostable URL shortener: one Rust binary, sharded-folder key/value store, API + web app + CLI. Runs hop.rip.
An NES CPU compiled into the weights of a stock Qwen3.5 model
A local coding agent with a web UI that builds itself while it runs, on Pi Durable
Examples, end-2-end tutorials and apps built using Liquid AI Foundational Models (LFM) and the LEAP SDK
A framework for building decision-model harnesses for any kind of task. Your program offers the model a short list of next steps, and the model picks one at a time until it can answer.