Agentic AI Designs an Analog Synthesizer
Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.
OpenAI’s GPT-6 Astra and Anthropic’s Claude Opus 5.5, released in September 2026, demonstrated markedly improved spatial reasoning and engineering capabilities, prompting one developer to test Claude Opus 5.5 via Claude Code on a from-scratch analog synthesizer design. Starting from a blank project folder and the JLCPCB free parts list, the agent proposed an MCU-controlled subtractive synthesizer built around a CH32V00x microcontroller, featuring a Juno-style digitally controlled oscillator, a Moog-style transistor ladder filter, and a voltage-controlled amplifier. The agent designed the circuit in KiCad 10, simulated it in ngspice, built its own component library from web-sourced models, and corrected the initial design based on simulation results, while also producing a Python behavioral model calibrated against the ngspice output to generate audio examples. The developer attributed success to a systematic, specification-first workflow rather than iterative trial-and-error editing, arguing that models generate each abstraction layer more reliably from a complete specification in one pass. The result matters as an early proof point that agentic AI can carry a hardware design from concept through simulation and documentation, and the next test is whether such end-to-end AI-designed circuits hold up in physical builds and scale beyond a single developer’s project.
Muhabbit's AI wrote this summary in its own words from Hacker News's report. It can get details wrong, so check the source.
Named in this story
GPT-6 Astra, Claude Opus 5.5, OpenAI, Anthropic, JLCPCB, CH32V00x