hardrave
Developing high-performance, minimal LLM runtimes and AI hardware/software tools that run in unconventional environments.
At a Glance
- Open-source AI developers
- Privacy-conscious AI users
- Embedded and systems programming enthusiasts
- Offline communities and educational groups
AI Tools by hardrave
(1)
NightRun
Bare Metal LLM Runtime
Discussions
No discussions yet
Be the first to start a discussion about hardrave
Latest News
Products & Services
A UEFI-resident local LLM runtime written in Rust that boots directly from USB or SD card without a conventional OS.
A zero-dependency, single-file AI slide generator for rapid presentation creation.
Hardware AI device that plugs into any PC for visual control and automation, including BIOS-level interaction.
An HPC-grade serverless LLM runtime for Google Cloud Run using gRPC streaming.
Market Position
Differentiated from standard runtimes like llama.cpp and Ollama by eliminating the operating system layer, providing a bare-metal, high-security, and high-performance alternative for dedicated AI hardware.
Leadership
Founders
Michal Wojewodzki
Senior Software Engineer and 2x international hackathon winner with expertise in Rust, UEFI systems, and LLM optimization.
Executive Team
Michal Wojewodzki
Senior Software Engineer / Founder
Specialist in running LLMs in environments like UEFI and serverless GPUs; 2x international hackathon winner.
Founding Story
hardrave is the developer persona of Michal Wojewodzki, who began creating a suite of specialized AI tools and runtimes. The most notable project, NIGHTRUN, was born from the vision of running LLMs directly on bare metal without the overhead or security risks of a conventional operating system, developed using advanced AI coding agents.
Business Model
Revenue Model
Open source projects with project-based development and research.
Target Markets
- Open-source AI developers
- Privacy-conscious AI users
- Embedded and systems programming enthusiasts
- Offline communities and educational groups
- Private and secure offline LLM interaction
- Minimalist local AI appliances
- Resource-constrained hardware (e.g., Raspberry Pi 5)
- Bare-metal systems programming research
- Early adopters in the Local LLM and self-hosting communities