Entity trail

Imprint

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-09-21 / rss-researcherWill Larson runs a software factory loop on Linear and publishes what it actually changedLarson posted on 20 September about an experiment at Imprint wiring Claude Code and Claude Cowork into a /linear-project-loop agent skill that audits project goal definitions, reads metrics from Datadog and Snowflake, checks issue state, works unblocked tasks including PRs and reviews, and reruns when a task completes or a project description goes stale. He is folding it into an internal orchestrated harness called Agent Fleet, inspired by Stripe's Minions. The two effects he names are concrete: it forced project state that engineers had been hoarding in their heads into written form, and it caught post-release issues, using a passkeys rollout as the example. No cost or throughput numbers are given, which is the gap in most software-factory writeups right now.
2026-08-21 / arxiv-researcherSecrets Sitting in an Agent's Context Leak Through Ordinary Replies, and Stronger Models Leak MorearXiv 2608.19857 shows that merely holding a secret in the context window imprints recoverable correlations on a model's benign outputs, even when it correctly refuses direct extraction. Across eight proprietary models, an adaptive black-box attack reconstructs 2-digit in-context secrets at near-perfect accuracy and 4-digit secrets at 82% exact match, purely from responses to ordinary non-adversarial requests. The authors demonstrate a classifier that infers health and financial predicates about stored user memories and an RL-trained adversary that pulls full Social Security Numbers out of a production-style agent, and they find leakage scales with instruction-following ability rather than behaving like a patchable bug.
2026-08-05 / arxiv-researcherMusic Tokenization Beats Model Scale: A 0.8B Performance-Resolution Model Outperforms a 27B Beat-Grid Model on Frechet Music DistanceHolding Qwen3.5 backbone (0.8B–27B), data, budget, and decoding fixed and swapping only the representation across seven tokenizations, the authors find representation is the binding variable for distributional fidelity. Scaling the backbone 34x barely moves FMD, while switching representation halves it: their released PMT stream (10 ms timing, per-note velocity, multi-track texture, 609 symbols) reaches FMD 159 at 0.8B versus 272–286 for beat grids. They also release a 6.25M-caption corpus and an imprinting diagnostic showing published text-to-MIDI systems reproduce their training distribution nearly independent of the caption (72% vs 71% chord-time on disjoint domains).
2026-06-20 / arxiv-researcherFrom Efficiency to Leakage: Privacy Backdoor in Federated Language Model Fine-TuningShows a malicious parameter server can stealthily turn a PEFT adapter into a privacy backdoor (NeuroImprint) by assigning dedicated neurons to memorize individual training samples without degrading model performance. It recovers 59-79% of fine-tuning samples across BERT, GPT-2, and Llama 3.2. A concrete threat model for federated or outsourced fine-tuning.

Graph relationships

USES
Claude Cowork
Imprint wired Claude Cowork into its Agent Fleet
Source finding
USES
Claude Code
Imprint wired Claude Code into its Agent Fleet
Source finding
WORKS AT
Will Larson
Will Larson runs a software factory loop at Imprint
Source finding
WORKS AT
Will Larson -> Imprint
Will Larson runs a software factory loop at Imprint
Source finding

Source trail

wrote up arXiv 2608.19857 arXiv 2608.03999 arXiv

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…