Analog-DB: An Agent-First Analog Integrated Circuit Database, From Blocks to Systems
Analog-db is an open-source, versioned database for analog integrated circuit designs. It uses a domain-specific language to capture designs as process-neutral topologies, reusable testbenches, and machine-readable datasheets. A parameterization scheme exposes functional sub-blocks and device sizes as named parameters with matching constraints. A schema-governed contract and queryable catalog allow AI design agents to discover and reuse circuits. Across the regulator corpus, all 23 circuit-kit bindings on three open kits meet their own recorded specification bands (typical corner, matched devices, no layout), and 10 of 23 meet a common class band. Seventeen of the 23 imported sizings failed their testbenches and closed under a gm/ID sizing loop driven by annotated sub-block roles, typically within one to three iterations.
Researchers introduced analog-db, an open-source, versioned database for sharing analog integrated circuit designs. It addresses challenges such as foundry NDAs and unreleased testbenches by using a domain-specific language that captures designs as process-neutral topologies, reusable testbenches, and machine-readable datasheets. The system includes a parameterization scheme for composability and retargetability, and a schema-governed contract with a queryable catalog for AI design agents. Evaluation on a regulator corpus showed all 23 circuit-kit bindings on three open kits met their own specification bands, 10 of 23 met a common class band, and 17 of 23 imported sizings that initially failed were corrected via a gm/ID sizing loop within one to three iterations.
The database uses a domain-specific language to unify topology, testbenches, and datasheets under one schema, enabling process-neutral design sharing. A parameterization scheme with named parameters and constraints supports circuit composability and retargetability. The gm/ID sizing loop, driven by annotated sub-block roles, successfully corrected 17 of 23 failed imported sizings, demonstrating automated design closure.
Analog-db addresses a key bottleneck in analog IC design: the difficulty of sharing and reusing designs due to NDAs and missing testbenches. By providing an open, versioned database with a queryable catalog, it could accelerate design reuse and enable AI-driven design automation in the analog domain, which has lagged digital design automation.
Analog-db could reduce time-to-market for analog IC designs by enabling reuse of verified, process-portable circuits. It may lower barriers for AI-assisted analog design, potentially disrupting traditional analog design workflows and creating opportunities for new EDA tools or design services.
Potential next signals include adoption of analog-db by design teams or EDA tool vendors, expansion of the design corpus beyond the regulator examples, integration with commercial process design kits, and publication of follow-up work on AI agents using the database for automated analog synthesis.