Breakthrough bloomnet May 1, 2026
x-sync CIB Detection: 8-algorithm ensemble, passive capture, knowle...
Manual investigation model: launch Chrom → Always-on passive capture of ALL GraphQL
Breakthroughs
Verified wins with real before/after metrics. Each breakthrough links back to the experiment chain and pitfalls that produced it.
31 total breakthroughs
23 from experiments
3 from skills
bloomnet (8)
Breakthrough bloomnet Apr 27, 2026
Vault skills dimension: 9 Rust scanners + repair sweep zeroed 42 st...
→
Breakthrough bloomnet Apr 27, 2026
MQI per-model baselines + multiclaude account detection: quality si...
→
Breakthrough bloomnet Apr 27, 2026
Agent MQI standalone repo: first public-facing artifact of bloomnet...
→
Breakthrough bloomnet Apr 27, 2026
Self-improving toolkit plugin: 3-axis taxonomy + 31 seed cases in o...
→
Breakthrough bloomnet Apr 19, 2026
Self-healing Claude Code hook install: symlink + wrapper preflight ...
~8-12 stderr lines per user turn from 17 → 0 stderr lines on sample Bash/Write/Edit
Breakthrough bloomnet Apr 8, 2026
Vault Knowledge Engine v1: 12,600 lines, 5 crates, 20 MCP tools, li...
Manual vault: grep-based search, LLM aud → 5 Rust crates, 12,600 LOC, 137+ tests, 7
Breakthrough bloomnet Mar 14, 2026
Three.js to Canvas 2D migration: eliminated WebGL issues, enabled L...
Three.js 3D scene with WebGL context tai → Canvas 2D with L-system botanical encodi
jobs-apply (12)
Breakthrough jobs-apply Apr 27, 2026
Apple code signing complete: Developer ID + notarization in one com...
→
Breakthrough jobs-apply Apr 26, 2026
Event-driven A/B system: 0% to fully autonomous in one session
0% operational: consent gate split-brain → Fully autonomous: 8 experiments running,
Breakthrough jobs-apply Apr 1, 2026
LinkedIn Easy Apply: 40% to 100% success rate in 10 runs over 18 days
40% submission rate, Run 1, 2026-03-15 → 100% submission rate (6/6), Run 10, 2026
Breakthrough jobs-apply Mar 30, 2026
Behavioral anti-detection skill: bot-mimicry pattern extracted from...
Account restricted, no systematic anti-d → Reusable skill: gaussian timing, reading
Breakthrough jobs-apply Mar 30, 2026
J-Score v2: three-layer fusion scoring, score clustering eliminated
Single-layer LLM scoring (Gemini Flash), → Three-layer fusion scoring, 592 tests, 3
Breakthrough jobs-apply Mar 24, 2026
LinkedIn anti-detection suite: account unrestricted, 83% session su...
Account restricted (Run 6, 2026-03-25). → Account unrestricted 2026-03-26. 83% ses
Breakthrough jobs-apply Mar 24, 2026
Multi-process coordinator: per-channel process isolation with auto-...
In-process ChannelRunLoop: cross-platfor → Per-channel child processes with WorkerS
Breakthrough jobs-apply Mar 23, 2026
Full marketing website built in 97 minutes (5 commits, 2,626 lines)
No public-facing website. Dashboard-only → Landing page, How It Works, Pricing, Wai
Breakthrough jobs-apply Mar 14, 2026
LinkedIn submission rate ratchet: 40% to 83% over 7 runs
40% submission success rate, Run 1 → 83% submission success rate, Run 7. 26 i
Breakthrough jobs-apply Mar 9, 2026
Direct channel: 98% submission success rate (48/49 applications)
Unknown success rate, no systematic meas → Direct 98% (48/49), Greenhouse 100% (10/
Breakthrough jobs-apply Mar 8, 2026
Infinite run coordinator: persistent autonomous job application engine
Single-run tool requiring manual restart → Persistent engine: per-channel state mac
Breakthrough jobs-apply Feb 25, 2026
AutoHunt: 15K-line job automation monorepo scaffolded in one day
No automation: manual job applications o → Full monorepo: 9 packages, 107 files, 10
oil (9)
Breakthrough oil Apr 13, 2026
Calibration-proper scoring eliminates Goodhart gaming: model probab...
v25.1 gamed: model 5-25x more bullish th → 260414b: MC forecast Day 46 median $97 (
Breakthrough oil Apr 3, 2026
Oil model v18.1 beats Polymarket by +2.16pp Brier score
Stale 10-day-old model, degenerate 0%/10 → R²=0.9134, MAPE=2.26%, beats Polymarket
Breakthrough oil Mar 30, 2026
Monte Carlo pricing engine v17.2: R²=0.975, MAE=$0.83, Brier +11pp ...
v12: OOS R² 0.8633, MAE $1.633, 0/5 hist → v17.2: OOS R² 0.975, MAE $0.833, Brier +
Breakthrough oil Mar 17, 2026
Oil v16 quant audit: +11.21pp Brier edge vs Polymarket, 82% win pro...
No external probabilistic validation: mo → +11.21pp Brier edge vs Polymarket, 82% w
Breakthrough oil Mar 17, 2026
Sell model: 100% trigger accuracy, zero false sells, CVaR99 -24.6%
No systematic sell discipline: manual ex → 11-trigger sell model, 100% accuracy, 0
Breakthrough oil Mar 17, 2026
Realtime consensus pipeline: autonomous hourly model updates, cumul...
Manual parameter updates, 2-4 hours per → Hourly autonomous updates. Cumulative v1
Breakthrough oil Mar 16, 2026
AR error correction: one-step MAE -33%, direction accuracy +6.6pp, ...
One-step MAE $1.63, direction accuracy 8 → One-step MAE $1.09 (-33%), direction acc
Breakthrough oil Mar 16, 2026
Sigmoid calibration: MAE -42%, autocorrelation eliminated
MAE $1.633, autocorrelation 0.1651 → MAE $0.951 (-42%), autocorrelation -0.00
Breakthrough oil Mar 15, 2026
Tail risk recalibration: OOS R² +4.98pp, tail probability >100% imp...
OOS R² 0.8633, P($120+) 15.4%, historica → OOS R² 0.9131, P($120+) 36.4%, historica
