Alex Bouchard

Forward-Deployed AI Lead

One person. Two consumer AI products live on the Apple App Store, three shipped. No handoffs. I take ambiguous AI products from decision to deployment — product, architecture, implementation, evals, native release, telemetry, and production outcomes — and I own the result.

01Shipped work

Live — Web + Apple App Store

Daily word game with a persistent generative character that reacts to real play. Deterministic code owns the puzzle, score, and limits; the model produces bounded character behavior from structured game context.

  • Server-owned rules prevent the generative layer from ever touching game truth — score, answer, and limits are unreachable from the model.
  • Character stays recognizable across guesses, hints, difficulty modes, wins, losses, fallbacks, and model changes.
  • Automated regression checks gate every release, including character-behavior changes.
  • Capacitor ships the public iOS app from the same product core as the web game — one implementation, two surfaces.
reacttypescriptgeminicapacitorsupabasevercel

That's My Best

thatsmybest.com

Live — Web + Apple App Store · concept to App Store in 14 days

Multimodal social game that turns social photo-grid screenshots into a playable friend quiz. The model proposes the answer key; code bounds what a wrong one can reach.

  • The model's guess ships as the answer key by default, so machine-checkable invariants bound it: no answers in the player payload, reveal only after a valid pick, an immutable seal, and scores only for players who actually played.
  • Privacy-preserving by construction: creator-supplied screenshots, independent per-photo verdicts, no facial recognition, no Meta API access, no staff photo review.
  • Early soft-launch generation cost measured at roughly 6–7¢ per completed quiz.
  • Scoped to objective visual memory over personality judgment, so the model never invents claims about a person.
multimodal aicapacitorserverlessweb + ios

Yapoleon's Court

court.yapoleon.com

Live — Web

Competitive AI game built so the model can't be gamed — structurally, not with prompt hardening. A player actively trying to manipulate an LLM, on a paid API, in front of a family audience.

  • The model never emits the score. It returns taste only — five bounded axes plus one in-voice line — and the favor delta is derived server-side by a pure function the model cannot see or reach.
  • Denial-of-wallet bounded five ways: 3-turn daily cap, per-IP and per-user rate limits, atomic concurrency slot, cached degrade mode, and a hard spend stop.
  • Deterministic red-line pre-filter runs before any model call, so flagged input costs zero tokens.
  • Prompt injection handled structurally — the player's reply rides as data being judged, never inside the system instruction.
reactgeminivercelsupabase

Open source — Python

Six synthetic executive lenses with competing incentives audit an AI-built operating plan that forecast ~$65.9M in profit. Rebuilt as a transparent, reproducible model when consensus alone failed to catch the arithmetic.

  • Reconciled a $3.397M gap between reported and fully loaded base-case expenses that survived six rounds of review.
  • 100,000 seeded Monte Carlo trials stress-testing sell-through, revenue execution, expense variance, weather, sponsorship, and delays.
  • Median reconstructed profit $1.22M; probability of reaching at least $5M: 10.5%.
  • Invariant tests block incompatible expense definitions from quietly surviving consensus.
pythonmonte carloregressioninvariant tests
02How I run models in production
Truth
Server-owned and invariant-bounded.
State
Deterministic, kept outside the model.
Safety
Untrusted content isolated as data, never instruction.
Reliability
Failure paths designed in advance, not discovered in production.
Economics
Cost treated as a product constraint with hard stops.
Verification
Release gates across the stack.
03Open source
04Distribution & measurement
  • Operate public web properties, treating technical SEO and generative-engine optimization as product infrastructure rather than marketing.
  • PostHog telemetry connecting model output to user behavior across web and native iOS — starts, submissions, completions, hints, shares, and purchases.
  • @YapoleonGreater on X run as a governed, near-autonomous publishing system that holds character voice inside defined boundaries.
05Before code — commercial real estate

Commercial Real Estate — Investment Sales

2012 – 2025
  • 85+ retail transactions representing roughly $400M in transaction volume as half of a two-person team.
  • Rebuilt a 5,700+ contact CRM and corrected approximately 1,900 mapping errors — ground-truth data work at production scale, before it was called that.
  • Thirteen years owning deal outcomes end to end: sourcing, underwriting, negotiation, and close.
06Stack & education
TypeScriptReactNext.jsPythonNodeSupabasePostgresVercelCapacitor / iOSGoogle GeminiMultimodal generationEvals & invariant testingPostHogTailwind

Education
B.B.A. Finance — Texas A&M University

Generate boldly. Validate cheaply. Kill ruthlessly. Scale what survives.