Sean Sooch

Incipit

Here begins a record of working systems, with commentary in the margins.

Medicine · the military · machines

SeanSooch

Working systems, with commentary

Thirteen working systems where medicine, the military, and frontier AI meet, built by a medical student and military officer with AI as the engineering tool. 150,906 lines across 247 files. Every one is live, and this page asks each host to answer while you read.

Written and built by Sean Sooch. This view set by Claude Opus 5.5, third in line, beside the Field Manual of Claude Fable 5.1 and the Personal Index of OpenAI Astra. One site, three views; the switch at the foot of the page keeps your place.

i

Prologue

Purpose, scope, and method

Purpose

This record documents five months of learning and building in spare hours. These projects were made possible by tools that lowered the technical barrier between an idea and working software. Others will bring different experience and build better things.

First deploy April 2026. Everything since, in evenings and weekends.

Scope

Everything listed is deployed and public. The page does not take that on faith: when it loads, your browser asks every host listed under Attestation to answer, and records what came back. The Works follow by domain, then the Writings, the lesser works, and the films. The Editions are the change record for the site itself.

A green mark means a host answered, not that its page is right. See the caution.

Method

Every system was specified by the author and written with Claude, then verified the way a clinician verifies a lab: against a source, by a second reader, and against the live deployment rather than the local copy. Medical content carries its citations. Scoring tables cite the governing publication. Where a source deck was wrong, the question was corrected and the correction logged.

NBStudy tools here are for exam preparation and are not medical advice. Calculators and field references are teaching aids; confirm with the governing guideline before use on a patient.

ii

The Works

Thirteen systems, by domain. Each name is a link.

In the margin beside each: what it holds, and whether its host answered this page.
Read
five works
Medicine
1Medicineasking

Step 2 CK Qbankrebuilt

Public study site over seven Anki-derived banks: 12,972 board-style questions and 12,357 guide facts across eight subjects, a UWorld-style test engine with tutor and exam modes, per-system performance, and three drill tools. Built by a chunk, author, independent-review, adversarial-triage pipeline that caught and corrected about 100 factual errors in the commercial source decks.

Open the question bank →
  • 12,972questions
  • 8subjects
  • 2-3review passes

step2-ck.pages.dev

2Medicine · QIasking

Patient Handoff Tracker

Full-stack I-PASS/SBAR handoff system built as an emergency medicine QI project: named-receiver read-back that only the receiver can acknowledge and verify, action items, Joint Commission compliance metrics with CSV export, and HIPAA-conscious audit logging over synthetic patients. Flask and SQLite on a free tier, so the first load wakes the server.

Open the tracker (about 30 s to wake) →
  • 22API endpoints
  • 8DB tables
  • Flask

patient-handoff-tracker.onrender.com

3Medicine · Cardiologyasking

Read This

Pattern-interpretation trainer: 28 procedurally generated rhythm strips at 25 mm/s with calipers and a live monitor, a fixed 8-step read of 14 twelve-lead cases with MI localization by territory, and an outside-in chest film read over synthesized radiographs and real films. Every case sourced.

Open the reading room →
  • 28rhythms
  • 8read steps
  • 100%sourced

seansooch.dev/ekg

4Medicine × Gameasking

MICRO TD

Tower defense where the body is the map: 16 guideline cases as levels, 34 real drugs as towers against 25 organisms, damage set by coverage and site penetration, resistance bred by antibiotic pressure, and organ toxicity, plus a one-patient Ward Rounds mode. TypeScript and Canvas, every fact sourced, 104 tests.

Play →
  • 16guideline cases
  • 34drugs
  • 25organisms

micro-td.pages.dev

5Medicine · Educationasking

AdmissionsAcademy

Founded by Paul Clancy to expand access to medical education: a free Pre-Med Essentials course, MCAT strategy, and application guidance for students from any background. Sean serves as Chief Technology Officer.

Visit →
  • Freepremed course
  • 1-on-1guidance
  • CTOSean's role

admissionsacademy.org

one work
Military
6Militaryasking

Warrior Ready

All-branch PT scorecard at 2026 standards: six services scored from their official charts with the source cited, per-service body composition screens, run pacer with lap splits, ruck planner, and score history. The build refuses to ship unless 72 spot checks read off the printed charts pass.

Open →
  • 6branches
  • 72chart spot checks
  • 0deps

warrior-ready.pages.dev

two works
Machines
7Machines · AI safetyasking

AGI Capability Tracker

Frontier capability ledger across 10 sectors, 57 models and dozens of benchmarks: a weighted composite score with its method published, verified ARC-AGI cells, a capability-milestones register, a capability-against-control safety ledger, and a gated-tier toggle for what the labs hold back. Updated September 2026.

Open the ledger →
  • 57models
  • 10sectors
  • 5milestones

differential-c57.pages.dev

8Machines × Medicineasking

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 and depth feedback, AHA 2020), Half-Life Tracker (caffeine, hydration, and BAC pharmacokinetics), and Pocket Telescope (live night-sky ID and star quizzes).

Run →
  • 4sims
  • 24decisions
  • 11endings

seansooch.dev/simulators

four works
Sports
9Sports · Fantasy footballasking

Edge Draft Terminal

Draft intelligence terminal for a 12-team half-PPR league: an 862-player board valued on a weighted seven-source composite and read against the Sleeper room's market price, weekly projections priced as points above replacement, Monte Carlo pick-path simulation in a worker, an in-season wire of injuries and market movers, and read-only replay of a live Sleeper draft room. Static PWA with CI-gated deploys and a test suite that also runs against production.

Draft →
  • 862players
  • 7sources
  • 21,761lines of tests

edge-7yj.pages.dev

10Sports · Hoopsasking

NCAAT Bracket

Archive of the finished 2026 men's tournament: all 67 games from a frozen ESPN snapshot, every first-edition model graded against the results, and an interactive bracket that scores your own picks. A sourced fact-check corrected 531 first-edition claims, all listed on the page.

Open →
  • 67games
  • 4models graded
  • 531corrections

ncaat-bracket-2026.pages.dev

11Sports · Golfasking

Masters 2026 Hub

Archive of the 2026 Masters as a hand-operated leaderboard: the final board for all 91 players with hole-by-hole cards from a frozen ESPN snapshot, an 18-hole yardage-book course guide, a searchable field, the champions' roll, and a Remotion film rendered from code. 137 corrections to the first edition are listed.

Open →
  • 91players
  • 18holes
  • 600frames

masters-2026.pages.dev

12Sports · Detroitasking

Detroit Sports Hub

Hub board for the Lions, Tigers, Pistons and Red Wings plus three trackers: a new Lions app (live down and distance on a drawn field, drive charts, win probability), Tigers with pitch-by-pitch tracking, Statcast and a 30-prospect farm system, and the Pistons' 14-68 to East #1 seed run with injury watch and a 2004 tribute.

Open →
  • 3trackers
  • 29tabs
  • 0deps

detroit-sports.pages.dev

one work
Humanities
13Humanities · Literature and philosophyasking

Great Books

Continuous-zoom library of 46 works across existentialism, political order, liberty, machines, knowledge, and reality.

Read →
  • 46books
  • 5shelves
  • 4depths

great-books.pages.dev

iii

Writings

Three papers and two reading rooms

The margin lists each paper's parts in the order it argues them.
IApril 2026Independent research

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.

Read on the AGI TrackerTechnical PDFPlain-language PDF

  1. Model riskdeception, unsafe knowledge, concealment
  2. System riskorchestration, tool use, parallel execution
  3. Deployment riskreal permissions, production access
  4. Access riskconcentration, cognitive inequality
  5. Governance riskweak audits, regulatory capture
  6. Civilizational riskcompounding failures across all layers
IIJune 2026Independent research

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.

Academic PDFSimple Mode PDFGame Theory Mode PDF

  1. Steerbuild inside guardrails and tripwires
  2. Speedthe race cannot be stopped, only won
  3. Stoppause until alignment is solved
  4. X-risklosing the future entirely
  5. S-riska future that continues, and hurts
  6. The gameone matrix, three readings, edit the payoffs
IIISeptember 2026Independent research

Two Stories About the Beginning

The Big Bang and a generated origin are the same kind of story: a starting condition plus a rule, with the evidence only ever reaching the rule. Four objections (tidiness, cheating, motive, proving too much) are taken seriously; two of them force the second story to change, and it survives as a testable rival rather than an evasion.

Read the paperThe two stories, side by sidePart II, stated in full

  1. Start + ruleevery origin story, and the evidence reaches only the rule
  2. Tidinessthe first story already carries the extra layer
  3. Cheatingno shortcuts; the saving is that the rule is short
  4. Motivedropped; the process only has to exist and run
  5. Last Thursdaya rival commits to something, an evasion to nothing
  6. The testsfinite grain, finite information, more room outside
two rooms
Reading rooms
asking

Reading Room

An annotated card catalogue of 64 AI papers, X threads and blog posts collected through April 2026, each with a technical summary, a plain-language version and a note on why it matters.

papers · posts · threads
reading-room.pages.dev
asking

Philosophy Room

One thesis: every great question converges on alignment, whether we can build minds that want what we want. Cosmology, consciousness, AI, identity, theology and game theory, each question read twice: as the classical problem and through the AI lens.

19 essays in 6 sections, about 30 minutes
philosophy-room.pages.dev
iv

Lesser works

Smaller experiments, same hands, same months

asking

Forge

Prompt-to-HTML site builder on the Anthropic streaming API, clone mode, component injector, live theme controls.

Claude API · vision
forge-8ce.pages.dev
asking

Hospital and Field Toolkit

Clinical calculators, antibiotic coverage matrix, differentials, lab reference with the CBC + CMP decoder, and TCCC MARCH: five tools, one hub.

Medicine · TCCC
seansooch.dev/toolkit
asking

Space Explorer

Live dashboard in eight sections over twelve public feeds, seven of them NASA's: APOD, the ISS over a map with its ground track, launch countdowns, EPIC Earth imagery and events, the Sun and its flares, near-Earth approaches, deep space, news.

NASA · live APIs
cic-dashboard.pages.dev
v

Films

Rendered from code, captured from live sites. Sound on.

Plate. The Field Manual (view F) in motion: the cover, the topology answering, the roll call, the record, the index. 1:20.
vi

Attestation

What is claimed, and what answered

Counts recounted 14 September 2026. Answers measured on every page load.
The counts
MeasureResultStandard
Lines of code150,906unique source files, generated data excluded; 21,761 of these are one project's test suite
Source files247html, js, css, python, typescript
Board-style questions12,9727 question banks, 8 subjects, every item independently reviewed, most a second time
Players and teams tracked1,106sports intel across 12 data sources
Hosts answering…of 18, measured from your browser now
Median round trip…milliseconds, edge-cached static hosts; the one server-rendered app is left out while it wakes
Months since first deploy5April to September 2026, evenings and weekends
The answers
SystemHostAnswerRound trip

CaveatThe ask is an opaque cross-origin fetch, so it can only report that a host answered, not that the answer was the right page. Cloudflare Pages serves a soft 200 for unknown paths. Read a green mark as "the lights are on", and click through for anything that matters.

vii

Editions

The change record, drawn as a family tree

One redesign per model class. Retired editions stay online.
The practice

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 is named for, then retired, not deleted, when its successor arrives. Put two side by side and you can see, in typography and judgment rather than benchmark scores, how far the frontier moved. Vol. 03 changed the practice: when the next class arrived it was not retired. The Field Manual stays current beside OpenAI Astra's Personal Index, and this Commentary joins them third, so Vol. 03 has three views drawn by three hands, and you can switch between them without losing your place.

A stemma is the family tree an editor draws of the surviving copies of a text. Each copy is a witness, named by a letter, its siglum.

Stemma
  1. ωThe work itselfThirteen systems, three papers, five films. Every edition below is a copy of this.
  2. Vol. 01The MagazineClaude Opus 4.7 · April 2026 · retired with honors · read the archive
  3. Vol. 02Triad × VitalsClaude Fable 5 · June 2026 · retired with honors · read the archive
  4. Vol. 03Three viewsOpenAI Astra + Claude Fable 5.1 + Claude Opus 5.5 · September 2026 · the text the three current views share
FThe Field ManualClaude Fable 5.1Vol. 03 · first in line. Numbered chapters, live readiness, a printed index. Read this view → AThe Personal IndexOpenAI AstraVol. 03 · second in line. A white exhibition catalogue: the name set huge, numbered editorial rows. Read this view →
Frontier leadOThe CommentaryClaude Opus 5.5Vol. 03 · third in line. A glossed edition: the work in the centre, commentary in the margins, one rubric red. You are reading this view.
Stemma. Solid line: descent. Vol. 01 and 02 are retired and kept online. Vol. 03 is the text the three current views share: F, A and O are its three views, in the order they joined; O is the frontier lead.
The record

Change record. Editions of this site, in order of issue.

EditionDesignDesigned byIssuedStatus
01The Magazine: dark editorial, sixteen sections, terminal metricsClaude Opus 4.7April 2026Retired with honorsarchive
02Triad × Vitals: three worlds, one chart, reactive inkClaude Fable 5June 2026Retired with honorsarchive
03Three views: the Field Manual, the Personal Index and the Commentary, switchable in placeOpenAI Astra + Claude Fable 5.1 + Claude Opus 5.5September 2026Currentyou are here
04To be designedThe next class-Awaiting release
Since Change 3

What changed beyond the design. The site is now three views of one text, switched from the bar at the foot of every page, which carries your place across, and each view's own record now lists all three. Every description was checked against the live apps on 23 September 2026 and corrected where an app had moved on; the corrections are one shared file, so all three views say the same thing. The Two Stories paper is set in each view. And the apps began keeping editions of their own: fifteen were redesigned from a blank page by Claude Opus 5.5 on 22 and 23 September 2026, and every retired edition stays online at its own address, listed below.

The apps' editions

App editions. Current edition of each app, and where its predecessor is kept.

AppCurrent editionDesigned byRetired edition
Step 2 CK QbankWorkroom 22 Sep 2026Claude Opus 5.5Edition 1
Patient Handoff TrackerChart Tab 22 Sep 2026Claude Opus 5.5Edition 1
Read ThisReading Room 22 Sep 2026Claude Opus 5.5Edition 1
MICRO TDOil Immersion 22 Sep 2026Claude Opus 5.5Edition 1
Hospital and Field ToolkitPocket Cards 22 Sep 2026Claude Opus 5.5Edition 1
Warrior ReadyScorecard 22 Sep 2026Claude Opus 5.5Edition 1
AGI Capability TrackerThe Ledger 16 Sep 2026Claude Fable 5.1Edition 2
SimulatorsDrawing Set 22 Sep 2026Claude Opus 5.5Edition 1
ForgeCasting Floor 22 Sep 2026Claude Opus 5.5Edition 1
Space ExplorerFlight Plan 22 Sep 2026Claude Opus 5.5Edition 1
Edge Draft TerminalEdition 1 2026earlier Claude modelsredesign queued
NCAAT BracketStudio 23 Sep 2026Claude Opus 5.5Edition 1
Masters 2026 HubBig Board 22 Sep 2026Claude Opus 5.5Edition 1
Detroit Sports HubWoodward 22 Sep 2026Claude Opus 5.5Edition 1
Great BooksTri-band 22 Sep 2026Claude Opus 5.5Edition 1
Reading RoomThe Stacks 22 Sep 2026Claude Opus 5.5Edition 1
Philosophy RoomBlack-Figure 22 Sep 2026Claude Opus 5.5Edition 1

Index

Every work, writing, film, and edition on this page. Press / to search it.

Explicit

This view was set by Claude Opus 5.5 in September 2026, third in line, beside the Field Manual of Claude Fable 5.1 and the Personal Index of OpenAI Astra. Three hands drew three views of one world. The work they describe is Sean's, who asked each of us to make it well, and put our names on it. ❧

Set in Alegreya, by Juan Pablo del Peral, and Cormorant Garamond, by Christian Thalmann.
Built as one static page. Nothing on it is a mockup.