Software development, game design, and AI research.

Consulting and client work through the Forge.

Forge with us
01 What we build

Three-layer brain architecture with task classification, model routing, persistent memory, and a full tool system. Plugin-driven extensibility from the ground up.

Python Plugin Architecture LLM Integration

Multi-dock architecture connecting coding tools to local inference. Crucible for direct GPU computation, MLX on Apple Silicon, llama.cpp on NVIDIA/AMD/CPU, with optional frontier API mooring for tasks that exceed local capability.

Python Go Crucible MLX llama.cpp

Parses GGUF model weights directly, runs transformer computation on GPU, generates text. No wrappers, no third-party inference servers.

Python CUDA GGUF
02 The foundation

The Flower Garden

We own every layer.

garden.grimvane.ai

Inference to storage. Every library built and owned by Grimvane.

03 The armory

Projects

04 The forge
Forge

Need something built?

Websites, apps, games, tools, automations. One person, full ownership, no handoffs.

forge.grimvane.com
05 Research
Research

Untrusted Weights

LLM security research. Risks of open-weight models in sensitive environments. Supply chain threats, prompt injection, model poisoning, and mitigation strategies. Two parts, twenty chapters, attack lab code.

About

Grimvane

One studio. No investors. Everything built from scratch, everything owned in full. AI assistants, puzzle games, security research, and client work — connected by the same standards and the same person behind each one.

Python AI/ML Game Engine
Access restricted