The Map

Commodity Futures · Learning Experiments

Monte Carlo pricing engine for WTI crude with calibration-proper dual-platform ratchet against Polymarket and Kalshi prediction markets.

Self-improving data models: 16 forward MC params auto-ratchet against live Polymarket + Kalshi via market-implied Brier + calibration penalty

Career Automation · Both Focuses

6-channel job automation with CDP browser control, behavioral adaptation, and J-Score v2 matching.

Self-improving scoring: J-Score v2 three-layer fusion learns from application outcomes

Agent Orchestration · Self-Improving Agents

Parallel Claude Code orchestrator with live API rate limit monitoring, spawn/kill UI, and Tauri v2 desktop.

Self-orchestrating compute: parallel agents with live resource monitoring and self-spawning

Developer Intelligence · Self-Improving Agents

Unified developer intelligence system: usage analytics, dynamic session context curation with adaptive forgetting, and multi-channel agent notifications.

Self-managing context + self-monitoring usage: scores, loads, and forgets context dynamically; visualizes the system watching itself; routes agent alerts across channels

Brand System · Learning Experiments

Brand system with R-generated design tokens, Paul Tol colorblind-safe palette, publication-quality chart themes.

Financial Tooling · Learning Experiments

MCP server for QuickBooks Online. 30-iteration audit across CPA, DE, and CFO personas.

Distribution Layer · Both Focuses

This site. Turns private experiments into public stories with real metrics. The distribution layer for everything else.

Agent Quality · Self-Improving Agents

Composite quality score that detects when your AI coding assistant is degrading. 24 metrics across thinking, research, execution, trust, throughput, and environment.

Self-monitoring quality: per-model baselines auto-calibrate, keyword sentiment tracks frustration as a leading indicator, real-time hooks warn when thinking depth drops

Shipping Automation · Learning Experiments

Shipping label automation and rate optimization.

Personality Distillation · Learning Experiments

Computable voice profile distilled from a decade of X/Twitter archive. Four measured lanes, 8 live posting rounds, brand-voice skill integrated into writing tasks.

Self-calibrating voice: archive-derived traits ratchet against posted-reply engagement

Personality Distillation · Learning Experiments

Second personality vector. Voice at the email register: how tone, formality, and structure shift when the channel changes. Gmail import pipeline feeds the same distillation engine as brand-voice.

Personality Distillation · Learning Experiments

Third personality vector. Visual personality extracted from the Apple Photos library using on-device ML metadata. Integrated as Step 15d of the BloomNet ingest.

Personality Distillation · Learning Experiments

Fourth personality vector. Consumption side of the flywheel: Meta feed and YouTube watch history ingested as vault frames, reconciled against the brand-voice production lanes. Project frame and two ingest experiments authored 2026-04-17; API and Takeout pipelines pending first run.