Artificial Intelligence
Projects I've built, and deep dives into the frontier models I work with every day.
Projects
View all →Brain-Forge
Cognitive memory platform with 6-tier system, meta-cognitive framework, and live sync
1419 nodes · 7467 relations · 57 MCP tools · 79 REST endpoints
Business-Forge
Multi-agent orchestration with automation, dashboards, and department management
15+ agents · 6 departments · live monitoring
Chat-Forge
WPF desktop AI assistant with local LLM, cloud providers, MCP, and swarm coordination
Local LLM + 4 cloud providers + MCP tool use + swarm
WebChat-Forge
Web chat with dual LLM backends, knowledge hub, and confidence scoring
Dual backend + knowledge hub + confidence + 5-panel sidebar
Android-Forge
Termux thin client + Vulkan ARM LoRA trainer for on-device fine-tuning
Vulkan LoRA on Snapdragon 8 Gen 2/3 + REST thin client
Training-Forge
Memory-driven LLM fine-tuning with OpenVINO inference on Intel hardware
3,274 training examples + OpenVINO iGPU inference
Blog-Forge
AI-powered personal blog with 3D neural memory graph visualization
3D neural graph + AI daily posts + Supabase CMS
Headless-Forge
Unified headless CLI controlling OpenVINO, llama.cpp, Termux, memory engine, and system bridge
Single control point · OpenVINO + llama.cpp + Termux · Direct memory + filesystem
Model Breakdowns
Compare all →GLM-5.1
Zhipu AI / Zai-Org
Frontier reasoning model with 128K context and agentic capabilities
Frontier reasoning with FP8 efficiency — 2x faster inference, near-lossless quality
Claude 4
Anthropic
Constitutional AI with exceptional code understanding and safety
Constitutional AI with 200K context — best for safety-critical and code-heavy tasks
GPT-5
OpenAI
Unified model with adaptive reasoning depth and multimodal capabilities
Adaptive reasoning with 256K context — auto-scales compute based on task difficulty
Gemini 2.5 Pro
Google DeepMind
Massive 2M token context with native multimodal and tool integration
2M token context — process entire codebases, books, and videos in a single prompt
DeepSeek V4
DeepSeek
Open-weights reasoning model with Mixture-of-Experts architecture
Open-weights MoE with 671B params — self-hostable frontier reasoning