Building generative AI creative pipelines — image, video, multimodal, and everything in between.

🎨 Image generation — BFL Flux workflows, ComfyUI custom nodes, and multimodal pipelines for AI image creation and manipulation

🎬 Video generation — agentic video workflows with Kling, Seedance, and other video models; using 3D as a control surface (not final render) to drive generative output

📝 Prompt engineering — crafting and refining prompts for LLMs and image/video models; multimodal (image-to-text, text-to-image) experimentation

🤖 AI automations — MCP-based agentic workflows, Claude + Blender pipelines, and tools that connect generative models to real production processes

🔬 Multimodal AI — bridging vision models, language models, and creative tooling

• 3D as a control layer for video model output (Blender → Kling / Seedance)
• Agentic camera control via MCP + Claude
• BFL Flux custom workflows and LoRA training
• Multimodal prompt design (vision ↔ language)
• Generative AI for brand and advertising production

Always experimenting. Always one step beyond.