AI Engineer specializing in Large Language Models, AI Agents, and Diffusion Models.
- Building and optimizing autonomous AI agent frameworks
- Developing browser-based agent interfaces
- Creating efficient agent orchestration systems
- Contributing to open-source agent development tools
- Implementing advanced fine-tuning techniques for modern LLMs
- Optimizing inference performance for large language models
- Working with models like Llama, Mistral, Phi, Qwen, and Gemma
- Developing post-training optimization techniques
- Implementing quantization techniques for diffusion models
- Optimizing memory usage and inference speed
- Working with SVDQuant and outlier absorption techniques
- Contributing to 4-bit model optimization
- Primary Languages:
- Python (AI/ML Development)
- CUDA (GPU Optimization)
- Frameworks & Tools:
- PyTorch
- Cog
- Hugging Face Transformers
- Custom Agent Frameworks
- Focus Areas:
- Agent Architecture
- Model Quantization
- Performance Optimization
- Inference Efficiency
- Advanced agent orchestration systems
- Memory-efficient LLM deployment
- Low-bit quantization techniques
- Multi-agent system architectures
- Edge device optimization
- Novel AI agent architectures
- LLM optimization research
- Diffusion model efficiency
- Open-source AI tooling
- Performance optimization techniques
- GitHub: @miike-ai
Building better AI tools, one commit at a time.