Title: Résumé — Alex Bouchard | Forward-Deployed AI Lead URL: https://midnightdev.dev/resume Description: Forward-deployed AI lead. Two consumer AI products live on the App Store, built solo end to end — model behavior, safety gates, reliability, cost controls, native release, and telemetry. Generated from the live page on 2026-10-22. Do not edit by hand; run `pnpm twins`. --- # Alex Bouchard Forward-Deployed AI Lead One person. Four shipped AI products, two of them live on the Apple App Store. 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. Houston, TX · open to relocation · [alex@midnightdev.dev](mailto:alex@midnightdev.dev) · [midnightdev.dev](https://midnightdev.dev/index.md) · [github.com/abouchard11](https://github.com/abouchard11) · [linkedin.com/in/alex-bouchard-ai](https://www.linkedin.com/in/alex-bouchard-ai) 01 Shipped work ## Dozen [midnightdev.dev/work/dozen](https://midnightdev.dev/work/dozen.md) Private product — graph and tests, not live GMV Agent marketplace between brands and 1k–50k TikTok / Instagram creators. The model drafts captions. Accept, post, and pay stay open until a human or an independent verifier closes them. - Twelve-node graph: intake → safety → assemble → offer → accept → draft → verify → revise → review → post → measure → pay. - Accept gate is dual — ≥60% and at least six seats. Accept stays closed until Stripe Checkout records the brand budget. Rejected seats stay in the offered denominator. - Post re-checks FTC disclosure fail-closed. Anomalous ER (3.2× or 0.15×) holds that seat's payout without closing pay. - 113 tests green as of September 1, 2026, production-imported, plus a graph linter that fails if the model owns an irreversible edge. - Stripe Checkout is wired fail-closed. Connect transfers and live TikTok/Instagram confirmation remain scaffolding. This line does not claim production GMV. tanstack start · postgres · invariant tests · closed-loop agents ## Yapword [yapword.com](https://yapword.com/) 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. - Three daily difficulty modes, themed games, contextual hints, and shareable challenges. - Capacitor ships the public iOS app from the same product core as the web game — one implementation, two surfaces. react · typescript · gemini · capacitor · supabase · vercel ## That's My Best [thatsmybest.com](https://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. - Answer-specific reactions, named friends, and timed group reveals drive the social loop. multimodal ai · capacitor · serverless · web + ios ## Yapoleon's Court [court.yapoleon.com](https://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. - Cached degrade mode serves an in-voice reaction with zero model calls under load; the meter never breaks and the character never breaks. react · gemini · vercel · supabase ## AI Boardroom Forecast Audit [github.com/abouchard11/ai-boardroom-forecast-audit](https://github.com/abouchard11/ai-boardroom-forecast-audit) 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. - Deterministic outputs, tests, evidence ledger, and limitations published publicly and labeled as reconstruction. python · monte carlo · regression · invariant tests 02 How 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. 03 Open source - [gemini-reliability-proxy](https://github.com/abouchard11) — Fallback chains and graceful degradation for model calls. - [llm-safety-gate](https://github.com/abouchard11) — Deterministic boundary isolating untrusted content from instructions. - [getzep/graphiti — PR #1637](https://github.com/getzep/graphiti/pull/1637) — Production validation on Neo4j + Gemini that caught a fallback path silently restoring a removed dependency; the fix shipped upstream. 04 Distribution & 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. 05 Before 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. 06 Stack & education TypeScript · React · Next.js · Python · Node · Supabase · Postgres · Vercel · Capacitor / iOS · Google Gemini · Multimodal generation · Evals & invariant testing · PostHog · Tailwind Education B.B.A. Finance — Texas A&M University Generate boldly. Validate cheaply. Kill ruthlessly. Scale what survives. Next step ## Want to talk it through? Book a 30-minute call. I will tell you plainly whether I can help and what I would test first. [Book a call](https://calendly.com/alex-midnightdev/30min?utm_source=midnightdev&utm_medium=site&utm_campaign=resume) [or email alex@midnightdev.dev](mailto:alex@midnightdev.dev)