Mauricio Martínez Langarica

AI-First Full-Stack Developer

I build complete products with AI — from idea to paying customer.

9 years running a real business. Today I design and build AI-assisted software already in production: multi-tenant SaaS, audit agents, and conversational agents serving customers 24/7.

Stack I work with

Next.jsReactTypeScriptNode.jsSupabase / PostgreSQLStripeAnthropic ClaudeGemini / OpenAIRAG / pgvectorPlaywrightVercel / Railway

From running a print shop to building with AI

I spent 9 years running Grupo Langamar Impresiones, a family B2B manufacturing business still in operation. That's where I learned to solve real problems with real constraints, not build for a portfolio. I bring that same standard to AI-assisted software today: every product I build has to hold up in production, not just look good in a demo.

  • 9 years running Grupo Langamar Impresiones, a family business still in operation
  • 3 in-house AI products in production: Camaleón, Matías, and Natalia
  • Certified AI Expert — Universidad Isabel I / BIG school, 150 hours
  • Founder of GenIA Labs, an AI audit and development agency for SMEs

Experience

From running a real business to building the software that runs it.

  1. FounderGenIA Labs

    January 2026 – Present · Mexico City
    • Founded a 360° Web Audit, AI Consulting, and Digital Development agency for Mexican SMEs.
    • Designed and built Camaleón, a multi-tenant SaaS for health clinics with 50 versioned migrations, 41 automated tests, and compliance with 4 Mexican regulations, in production.
    • Built Matías, a web-audit agent with 44 parallel analysis modules, 94 test files, and a mandatory human-review gate.
    • Deployed Natalia, a WhatsApp conversational agent running 24/7 with RAG (pgvector) and anti-hallucination guardrails, qualifying B2B leads in production.
  2. DirectorGrupo Langamar Impresiones

    2017 – 2026 · Mexico City
    • Led team, client, and supplier management at a family B2B manufacturing business in continuous operation.
    • Directed the business's digital marketing strategy.
    • Ran the company for 9 years, applying real-world business judgment to every technical and operational decision.

Stack

What I actually build with in production — not an aspirational list.

Frontend

Next.jsReactTypeScript

Backend & data

Node.js / ExpressSupabase / PostgreSQL / pgvectorStripe / Stripe Connect

AI

Anthropic ClaudeGemini / OpenAIRAG and AI agents

Tooling & deploy

Playwright / VitestVercel / Railway

In-depth projects

Three products in production, from architecture to regulatory compliance.

Camaleón

A multi-tenant SaaS platform for health clinics: a universal console with industry "cartridges." The Nutrition cartridge is already in production, on an architecture built to add more industries (e.g. dental) without rebuilding the core.

versioned migrations
50

versioned migrations

components
177+

components

automated tests
41+

automated tests

Mexican regulations met
4

Mexican regulations met

Next.js 16React 19Supabase (PostgreSQL, Auth, Storage)StripeResend

Matías

An internal GenIA Labs agent that audits websites across 15+ dimensions (legal, SEO, GEO, UX, performance, conversion, e-commerce, PWA, accessibility) in a single pass, with evidence and mandatory human review before delivery.

parallel analysis modules
44

parallel analysis modules

test files
94

test files

documented REST endpoints
38

documented REST endpoints

CI coverage gate
≥65%

CI coverage gate

Python 3.11FastAPIPlaywrightBeautifulSoupNext.js 16

Natalia

A conversational agent that handles WhatsApp for Grupo Langamar Impresiones: qualifies B2B leads, collects quote briefs, and escalates to the human director — live in production, running 24/7.

running in production
24/7

running in production

humanized reply delay
2–4 min

humanized reply delay

from the director → bot locks out
3 msj

from the director → bot locks out

Node.js / TypeScriptExpressClaude Haiku (Anthropic)OpenAI EmbeddingsSupabase + pgvector

Other projects

More work built with the same standards.

Pide Menú

A restaurant SaaS: a digital menu with configurable options, radius-based delivery zones (geocoding), optional Uber Direct integration, a diner-facing PWA, and LFPDPPP compliance. Subscription model with Pro, Lite, and Café plans.

Next.jsSupabaseStripe ConnectUber Direct (opcional)

Grupo Langamar Impresiones

A B2B corporate site for the family print shop: a full technical SEO/GEO strategy (schema markup, atomic answer capsules for AI Overviews, 23 preserved 301 redirects through the migration), 4 buyer personas with specific objections, and a documented conversion funnel with business and technical KPIs.

Next.jsSchema markupSEO/GEO técnico

GenIA Labs

My own agency's site (dogfooding): Next.js with i18n (ES/EN), a real GEO strategy (llms.txt, llms-full.txt, explicit rules for 9 AI bots in robots.txt, Q&A blocks for citation by generative engines), and 3 in-house AI agents documented with explicit limits on what they must not invent.

Next.jsi18n (ES/EN)llms.txt / GEO

Looking for someone who actually builds with AI?

Reach out directly. I answer myself.