Working notes on Piper. Shared copies — see the note at the top of each.
Piper: Architecture and Empirical Record of a Persistent, Self-Directed Language-Model Agent Succession, delegate-only action, held-out instruments, motivation, ideation and self-improvement, each with its evidence and what it does not yet show — v2.0.1, 17 Oct 2026 Foundation: A Reliable System Around an Unreliable Intelligence The machinery around the agent and what it has measurably achieved: handovers, workers and gates, instruments, money, records, claim checkers, boundaries and a move to a new machine — v1.3.0, 17 Oct 2026 The Salience Engine: from motivational pressure to a working attention system The design, private tests, a week on trial, the switch to live, and nine days of the loop running on its own — v1.3.0, 15 Oct 2026 Engineered Serendipity: a goal-conditioned stochastic ideation architecture Goal-conditioned random stimuli (kaleidoscope, Mandelbrot, inkblot) feeding Piper's ideation, and how she has used it live. A blank image does about as well as a real one; five improvements built — v1.6.0, 15 Oct 2026 Piper: Recursive Self-Improvement — Working Measurement & Engineering Paper What would count as evidence that Piper improves the machinery that improves her, checked against her own records. No amplification shown yet — v1.0.25, 20 Oct 2026 Piper: Adaptive Personality & Social Memory A build specification for the 38 agency traits: learned baselines, fading reactions, per-entity memory and habituation, with starting values computed from her own record, a social-memory core and a card switched on 17 Oct, its first two days live, and the follow-ups built — v1.0.7, 19 Oct 2026