nicolas mendez gonzaleznmg

Open to AI Engineering & Project Manager roles — Munich · Remote

Nicolas Mendez Gonzalez

I build AI systems that survive contact with reality.

Fifteen years engineering the networks behind connected cars, eSIMs and IoT taught me one thing: assume it breaks, prove it works. I now apply that discipline to LLM applications, agentic systems and automation.

See the work Get in touch
Portrait of Nicolas Mendez Gonzalez
Engineering experience
15+ years
AI apps shipped & live
3 public + private work
Certified
PMP project management
Working languages
EN · DE · ES

01 — Selected work

Built to be used, not just demoed

Every project answers the same three questions: what problem does it solve, how was it engineered, and what came out of it.

Turing College · AI Engineering

Flagship

Digital Marketing Assistant

An AI marketing consultant for small businesses: it answers from a curated marketing knowledge base with citations and manages Google Ads and Meta Ads campaigns — with a human approval gate on every write.

  • problemSmall businesses run their own paid campaigns without a marketing team, and letting an AI touch live ad budgets is only acceptable if it cannot spend money unsupervised.
  • approachA LangChain/LangGraph agent over OpenRouter models with hybrid RAG (pgvector + BM25, reciprocal-rank fusion) for cited answers, where every write tool returns a proposal that only human approval can execute — backed by budget guardrails enforced in code, not in the prompt, and a full audit trail.
  • outcomeA live multi-user app — Next.js frontend on Vercel over a FastAPI backend on Cloud Run — covering campaign management, a GA4 / Search Console / Shopify metrics dashboard, graded business audit reports, and a nightly sync that feeds anomaly alerts and email digests.
  • LangGraph
  • Hybrid RAG
  • FastAPI
  • Next.js
  • Supabase

Live app ↗

Turing College · AI Engineering

Interview Practice App

An LLM-powered app for practising job interviews: AI-generated questions tailored to the role, with structured feedback on every answer.

  • problemInterview practice usually needs another person — and generic question lists don't adapt to the role you're applying for.
  • approachPrompt-engineered question generation and answer evaluation, shipped as a live web app as part of the Turing College AI Engineering course.
  • outcomeDeployed as a live web app on Vercel: role-specific questions and structured, criteria-based feedback on every answer, usable for real interview preparation end to end.
  • LLM
  • Prompt engineering
  • Next.js
  • Vercel

Live app ↗

More projects — side projects & tooling

Some projects are private — I'm happy to walk through them in a conversation.

02 — Why me

What fifteen years of telecom buys you in AI

Most AI demos die in production. My whole career was making complex systems survive there.

Systems thinking

Connectivity only works when every layer — SIM profile, network, backend, vehicle — agrees. I design agent pipelines the same way: explicit contracts, defined hand-offs, no magic in between.

A tester's skepticism

As a test manager, my job was to break things before customers did. I treat LLM output the same way: measure it, validate it, never take it on faith.

Delivery, not just code

PMP-certified project leadership: requirements, architecture, people and customers. I've won bids, led teams and shipped on deadlines — and I bring that to AI work.

03 — Track record

From live networks to AI systems

During my professional career I moved from deep technical roles into test and project leadership — always in demanding, real-time environments.

  1. 2026 — Present

    AI Engineering

    Turing College

    Full-time deep dive into agentic systems, LLM applications, local AI and automation — building and shipping a project every sprint.

  2. Most recently

    Project Development Manager

    Accenture · Munich — automotive connectivity

    Led system architecture and integration projects for automotive connectivity solutions: designing and integrating systems, plus people leadership, requirements management, and bid preparation (offers and presentation material).

  3. Earlier

    Functional Owner

    eSIM / IoT systems · during my professional career

    End-to-end responsibility for system functions: requirements, specification and validation across the eSIM/IoT stack.

  4. Earlier

    Test Manager

    Automotive connectivity / eSIM · during my professional career

    Owned test strategy and execution for connectivity systems — planning, coordinating and validating across suppliers and in-car environments.

  5. Career start

    Senior Customer Support Engineer

    Mobile networks · during my professional career

    Front line of live mobile network operations — diagnosing and resolving complex issues directly with demanding customers.

04 — Toolbox

Skills

AI Engineering

  • LLM applications
  • Agentic systems
  • LangGraph
  • Prompt engineering
  • Local LLMs
  • Automation

Telecom & Connectivity

  • Mobile networks
  • eSIM
  • IoT systems
  • Automotive connectivity

Testing & Validation

  • Test management
  • Functional ownership
  • Validation strategy
  • In-vehicle testing

Leadership

  • Project management (PMP)
  • People leadership
  • Requirements management
  • System architecture
  • Customer engagement
  • Bid preparation

05 — The person

About

Nicolas Mendez Gonzalez is a Telecommunications Engineer (Universidad de Sevilla) turned AI Engineer, based in Munich, Germany. During 15+ years in mobile networks, eSIM/IoT and automotive connectivity he grew through successive positions — Senior Customer Support Engineer, Test Manager, Functional Owner and, most recently, Project Development Manager — leading system architecture and integration projects. Since 2026 he has been training as an AI Engineer at Turing College, building LLM applications, agentic systems, local AI and automation. He is PMP-certified and works in English, German and Spanish.

I've spent my career where failure is expensive and visible. Along the way I led projects and people, owned system architecture and requirements, and helped win work through bids and proposals. I'm as comfortable in front of a customer as in front of a terminal.

At Turing College I ship a working project every sprint. Same engineer, new stack: I still assume things break, and I still prove they work.

Location
Munich, Germany
Languages
Spanish (native) · English (business) · German (fluent)
Certification
PMP — Project Management Professional (valid to Apr 2029)
Education
Telecommunications Engineering, Universidad de Sevilla
AI Engineering, Turing College (2026–present)

06 — Next step

Let's build something that works

Open to AI engineering roles and conversations about agentic systems, connectivity and project leadership — in English, Spanish or German.

nicolas.mendez.gonzalez@gmail.com