About the service
I develop real-time audio programs in Pure Data (Pd) — a visual dataflow language on which installations, tools and stage systems are built. If you have a sound idea that needs to work reliably on stage or in a gallery, I turn it into a neat, clearly structured patch that other people can also use.
What I do: custom synthesis and processing patches (granular, additive, subtractive, FM, physical modeling), signal graphs for spatial and multichannel sound, generative and algorithmic music engines, interactive installations that respond to sensors, cameras, MIDI or OSC. I write and compile external objects in C when a patch needs performance or DSP unavailable in standard objects, and assemble everything into standalone builds so the client can run the project without installing Pd.
How I work: I keep the signal path readable through subpatches and abstractions, comment complex nodes and provide brief instructions so you can later change parameters yourself. I check CPU load, denormals, voice distribution without clicks and clean gain staging so the work holds up through a long exhibition or full concert. I connect hardware controllers, DAW via Jack or virtual MIDI, video through GEM and network messages via Open Sound Control.
Typical projects: responsive sound landscape for an exhibition, live instrument for a specific performer, control bridge from a browser into Pd, multichannel diffusion patch or a teaching example breaking down DSP technique step by step. Send a description, reference or sketch — I'll suggest the structure, timeline and fixed scope. You'll get editable patch files, externals with source code and notes on tuning for your machine and audio interface.
What I do: custom synthesis and processing patches (granular, additive, subtractive, FM, physical modeling), signal graphs for spatial and multichannel sound, generative and algorithmic music engines, interactive installations that respond to sensors, cameras, MIDI or OSC. I write and compile external objects in C when a patch needs performance or DSP unavailable in standard objects, and assemble everything into standalone builds so the client can run the project without installing Pd.
How I work: I keep the signal path readable through subpatches and abstractions, comment complex nodes and provide brief instructions so you can later change parameters yourself. I check CPU load, denormals, voice distribution without clicks and clean gain staging so the work holds up through a long exhibition or full concert. I connect hardware controllers, DAW via Jack or virtual MIDI, video through GEM and network messages via Open Sound Control.
Typical projects: responsive sound landscape for an exhibition, live instrument for a specific performer, control bridge from a browser into Pd, multichannel diffusion patch or a teaching example breaking down DSP technique step by step. Send a description, reference or sketch — I'll suggest the structure, timeline and fixed scope. You'll get editable patch files, externals with source code and notes on tuning for your machine and audio interface.
