Siddharth Ahuja is a designer, engineer, and creator of Blender MCP, an open-source integration that lets AI assistants operate professional 3D software through natural language. His work makes sophisticated creative tools accessible to people who know what they want to create but have not mastered complex specialist interfaces.
Ahuja developed Yoguide, a Copenhagen Institute of Interaction Design project that used smartphone-based computer vision to help blind and visually impaired people practice yoga independently. It received a UX Design Award in 2020. He subsequently worked as an entrepreneur-in-residence at LEGO Ventures and, from 2021 to 2024, co-founded and ran Prophecy, a product consultancy whose clients included Loop Health and Headout.
In 2025, Ahuja introduced Blender MCP, applying his experience in design, prototyping, and engineering to Blender’s steep learning curve. Because AI assistants can write code and Blender can execute scripts, he built an integration allowing users to describe scenes, objects, lighting, and animation conversationally.
Natural-language 3D creation: Blender MCP connects clients such as Claude and Cursor to Blender through the Model Context Protocol and a custom add-on. Integrations with Sketchfab, Poly Haven, and Hyper3D Rodin enable assistants to retrieve or generate assets and assemble complete scenes.
Lean, differentiated agent tools: Ahuja found that overlapping capabilities make models unreliable at selecting the correct action. His approach favors clearly distinct tools organized around the creator’s intended outcome, treating orchestration as a product-design challenge.
Cross-application creative orchestration: His Ableton MCP extends the same architecture to music production. Combining both integrations, he demonstrated an assistant creating a dragon scene, adjusting its lighting, and generating a corresponding soundtrack across separate applications.
Ahuja envisions creators directing games, films, and music through their artistic intent while AI coordinates specialized software and external services. He also acknowledges that language models still struggle with three-dimensional spatial reasoning and that generated results remain uneven.
Blender’s scripting interface lets a language model turn a scene description into operations. Siddharth Ahuja’s experiments show why distinct tools matter, and how the same approach can connect generated objects, animation, and music.
▶ Watch ↗