Sweepme logo

Autonomous Instrument Control

Let AI run your bench.

Operate your instruments with AI in the loop. You and your AI agent, hands on the hardware in real time. Be among the first.
Join the waiting listTalk to us

Example

Prompt your bench

Ask in plain language. The agent operates the instruments, evaluates the data and hands you the plot and the numbers.

Generator output

Measure the output of the signal generator at 100, 200, 400 and 800 Hz and plot it.

Signal generator output as measured on the scope at 100, 200, 400 and 800 Hz: four stacked sine traces over 40 ms, each swinging between −0.3 V and 0.7 V around the 0.2 V offset, aligned so that all four peak together at 0, 10, 20, 30 and 40 ms.
Recorded session
  • Set a 0.5 V sine with 0.2 V offset
  • Recorded 40 ms per frequency on the scope
  • Fitted amplitude, offset and phase, aligned all maxima at t = 0

0.499 V amplitude, 0.201 V offset

Set to 0.5 V and 0.2 V, the same at all four frequencies. Fit residual 0.7 mV.

Result in a few minutes

Demo bench: a Red Pitaya STEMlab as signal generator and oscilloscope, driven by its SweepMe! drivers.

An agent at the bench, not a script generator

Real instruments, live

Ports stay open for the whole session. The agent sends commands and reads values back from the hardware on your bench.

It decides the next step

Hypothesis, trial, error, retrial, until something works. No sequencer run written for every guess.

Every session on record

Commands and values are protocolled, so a run can be reviewed, handed over and repeated.

How it works

The agent talks to your bench over MCP, the Model Context Protocol.
SweepMe! is the host. It owns the drivers, the ports, the folders and the safety rules, and publishes a bounded surface of your bench to whatever AI is connected.

Made for bring-up and debugging, from ASICs to boards to any DUT, where the next step depends on the last measurement. The agent runs that loop. Control and the final interpretation stay with you.

Safety by design

Limits and permitted actions live in SweepMe!, not in a prompt.

300+ instruments ready

The feature builds on the SweepMe! driver library.

Next to the sequencer

Explore command by command, then keep what worked as a setting.

From first prompt to repeatable test

  1. 01

    Ask in plain language

    Describe the measurement, or the fault you are chasing.

  2. 02

    The agent measures

    It runs your instruments within the limits set in SweepMe! and evaluates every result.

  3. 03

    Take over any time

    Switch between manual and AI control as you like, on the same instruments.

  4. 04

    Define tests, re-run them

    Keep what worked as a test and re-run it whenever needed.

Features

What it does today, and what comes next

Available now

  • Control instruments from different manufacturers in one tool, by AI agent or by hand
  • Record every command of a session
  • Review earlier sessions in the session overview
  • Set limits and safety guardrails
  • Share sessions with colleagues via export and import
  • Test step by step: instruments keep their last state, so you plan the next action from there
  • Turn what worked into robust, ready-to-use test procedures
  • Evaluate data and create plots with analysis scripts
  • Let the agent write instrument drivers
  • Let the agent run experiments
  • Keep usage knowledge, so future sessions run more efficiently
  • Connect any AI agent, such as OpenAI Codex, Claude Code or Cursor

Planned

  • Let the agent create reports and test flows
  • Load tests into the SweepMe! sequencer and combine them with other instruments

Be part of it, and benefit first

We are opening Autonomous Instrument Control to a small group of early testers. Sign up and we will get in touch before the public release.