← @bgSponge

deepfates/imp: declarative self-improving language-model programs for Elixir 😇

Summary

deepfates' Imp is a DSPy port to the BEAM for declarative, self-improving language-model programs in Elixir.

  1. Signatures replace hand-written prompts.

    Imp builds the prompt from the signature, checks the reply, and returns typed fields or an error.

    You never write a prompt or a parser.

  2. Optimizers improve measurable programs.

    Labeled train, val, and test sets plus a metric drive GEPA, MIPROv2, SIMBA, few-shot bootstraps, and weight-training paths.

  3. Agents are OTP processes.

    Imp.start_run supervises a program so you can authorize tool calls, watch events, stop it, and tie its lifetime to the starter process.

    Declarative, self-improving language-model programs for Elixir.

  4. The same optimizers steer agents.

    GEPA can reflect on whole agent runs, and Optimize Anything can rewrite scorable tool descriptions or JSON.

    GEPA runs the program, reads where it failed, and rewrites its instructions.

    The result is a new program whose instructions and examples you can read, save as JSON, and review as a diff.

  5. 0.5 is experimental.

    The first Hex release may still change its API, and the README says its optimizers still need large-scale benchmarking.

Read the original ↗Save to your Sponge

Summarized with Sponge. Quotes are the author’s own words.

More from @bg

All readings →