Vol. 02 · 2026 · ten tools, three worlds, one builder

Ten shipped tools built solo at the intersection of all three, by a medical student and military officer using AI as a force multiplier. 52,417 lines in 27 days, every project live in production.

↑ hover a world to tint the page · click to filter the work

Launch film · Vol. 02 in motion 02:02 · sound on

Portfolio vitals

collected 04/2026 · live · reviewed at every model release
AnalyteResultReference range
Lines of code52,417H0-5000
Functions shipped845H-
Medical questions2,245H0-200
Players & teams tracked244sports intel, 5 data sources
Live deployments10all systems nominal
Days to build27L90-365 (traditional team)
Formal CS degrees0within expected limits

The work

Medicine

Step 2 CK Engine

Adaptive study platform: 2,245 questions and 1,211 flashcards across 351 topics, SM-2 spaced repetition, diagnostic surgery game, performance heatmaps, and a custom OCR ingest pipeline.

4,074 LOC2,245 Qs351 topics LiveOpen → Medicine · QI

Patient Handoff Tracker

Full-stack I-PASS/SBAR handoff system built for a real emergency department: Joint Commission compliance metrics, HIPAA-conscious audit logging, role-based access, and a QI dashboard.

20 endpoints7 DB tablesFlask Live · ~30s cold startOpen → Medicine · Cardiology

Read This

Pattern-interpretation trainer: 28 procedurally generated rhythms, a fixed 8-step 12-lead read with MI localization by territory, and an outside-in CXR module, all calibrated to 25 mm/s, all sourced.

28 rhythms8 read steps100% sourced LiveOpen → Medicine × Game

MICRO TD

Tower defense that enforces the drug-bug matrix: 16 pathogens, 25 antibiotics, resistance evolution on wrong-drug hits, synergy combos, and a Supabase leaderboard. React 19 + Canvas at 60fps.

16 pathogens25 drugs60 fps Live · best on desktopPlay → Military

Warrior Ready

All-branch PT command center at 2026 standards: six score calculators, body composition, run pacer with SVG splits, ruck planner, and score history. 52 scoring claims verified against official publications.

6 branches52 claims verified0 deps LiveOpen → Machines · AI Safety

AGI Capability Tracker

Real-time dashboard across 8 capability domains, 25+ frontier models, and dozens of benchmarks, composite AGI score, safety scatter plot, automated three-source data pipeline, and a restricted Mythos Mode.

25+ models8 domains3 pipelines LiveOpen → Machines × Medicine

Simulators

Four sims, one umbrella: the Alignment Simulator (run a frontier lab, 12 decisions, 6 endings, built from the Risk Stack paper), Code Blue (spacebar CPR with live rate/depth feedback, AHA 2020), Half-Life Tracker (caffeine, hydration, and BAC pharmacokinetics), and Pocket Telescope (live night-sky ID and star quizzes).

4 sims24 decisions11 endings LiveRun → Sports · Hoops

NCAAT Bracket

64-team interactive bracket aggregating four analytics models, KenPom, adjusted net ratings, advanced projections, with composable X-factor modes for injuries, guards, and 3-point shooting. Offline-capable PWA.

64 teams4 models3 X-factors LiveOpen → Sports · Golf

Masters 2026 Hub

Tournament companion with live leaderboard, an 18-hole Augusta course guide, a searchable 92-player field, champions gallery with Ken Burns animations, plus a Remotion renderer producing broadcast video from code.

92 players18 holes600 frames LiveOpen → Sports · Detroit

Detroit Sports Hub

Two broadcast-grade trackers in iframe isolation: Tigers with pitch-by-pitch tracking, Statcast, and a 30-prospect farm system; Pistons covering the 14-68 → #1 seed run with injury watch and a 2004 tribute.

14,776 LOC21 tabs0 deps LiveOpen →

Research & writing

2 papers · 2 rooms · 19 essays
Machines · Independent research · April 2026

The Frontier AI Risk Stack

Frontier AI risk is best understood not as a single category but as a stack of six interacting layers. The most serious dangers arise from the way failures compound across layers, a structured analytical tool for policymakers, developers, and civil society.

01
Model Risk
deception, unsafe knowledge, concealment
02
System Risk
orchestration, tool use, parallel execution
03
Deployment Risk
real permissions, production access
04
Access Risk
concentration, cognitive inequality
05
Governance Risk
weak audits, regulatory capture
06
Civilizational Risk
compounding failures across all layers
Read on AGI Tracker Technical PDF Plain-language PDF
Machines · Independent research · June 2026

Three Roads From Here

The three positions on frontier AI (safety through development, acceleration, pause) characterized on their strongest terms, with existential and suffering risk treated as separate axes that do not always move together. The central frame: the camps are reading three different payoff matrices for the same game.

01
Steer
build inside guardrails and tripwires
02
Speed
the race cannot be stopped, only won
03
Stop
pause until alignment is solved
04
X-risk
losing the future entirely
05
S-risk
a future that continues, and hurts
06
The game
one matrix, three readings, edit the payoffs
Academic PDF Simple Mode PDF Game Theory Mode PDF
Annotated reading

Reading Room

Papers, posts, and threads I'm actively working through, each with a technical summary and a note on why it matters.

papers · tweets · blogsEnter → 19 essays · 6 sections

Philosophy Room

One thesis: alignment is the most consequential question in human history. The case is built from cosmology, consciousness, identity, theology, and game theory.

~30 min readEnter →

Explorables

smaller experiments, same period
Forge Prompt-to-HTML site builder on the Anthropic streaming API, clone mode, component injector, live theme controls. Claude API · VisionOpen → Hospital & Field Toolkit Clinical calculators, antibiotic coverage matrix, differentials, lab reference, and TCCC MARCH, five tools, one hub. Medicine · TCCCOpen → Space Explorer Live dashboard on 20+ NASA APIs: APOD, EPIC Blue Marble, solar observations, ISS telemetry, asteroid proximity. NASA · Live APIsExplore → Pika: AI Meeting Agent An autonomous agent that joins live Google Meet calls, listens, and participates in real time. AI Agents · DemoWatch → Skubal Hype Video A broadcast-quality hype reel for Tarik Skubal generated entirely from React components and frame math. Remotion · SportsWatch →
Built with
  • Claude Code
  • Anthropic API
  • Multi-agent architecture
  • MCP stack
  • Cloudflare Pages
  • Flask + SQLite
  • React 19
  • Remotion
  • Canvas
  • Capacitor
  • Vanilla JS at scale

Assessment & plan

These projects were built by one person with no formal CS degree, in free time after clinical rotations and between study sessions. That alone says more about where the tools are heading than anything about me individually.

Capabilities are compounding faster than any individual can fully track, which is exactly why I track them. The professionals who thrive won't be the ones resisting AI or passively consuming it. They'll be the ones operating fluently in a dual ecosystem, human and machine in concert. The bottleneck is no longer technical skill. It's imagination, domain knowledge, and the willingness to build.

Disposition: vigilance.  Electronically signed: S. SOOCH

Editions

one redesign per model class · the site is the benchmark

This site is rebuilt from the ground up at every new Anthropic model class, as a personal benchmark of frontier capability. Each edition is designed end-to-end by the model it's named for, then retired, not deleted, when its successor arrives. The archive is the measurement: put two volumes side by side and you can see, in typography and judgment rather than benchmark scores, how far the frontier moved.

VolumeEditionDesigned byShippedStatus
Vol. 01 The Magazine: dark editorial, sixteen sections, terminal metrics Claude Opus 4.7 April 2026 Retired with honors View archive →
Vol. 02 Triad × Vitals: three worlds, one chart, reactive ink Claude Fable 5 June 2026 Current You are here
Vol. 03 To be designed The next class - Awaiting release