You're seeing this page as if you were . The main menu is still yours, though. Exit from immersion
Matthieu D.MD

Matthieu D.

AI & Automation Engineer n8n · LangChain · Python

400 €/día
Madrid, ES
3-7 años

Tiempo medio de respuesta: 1h

Acerca de Matthieu

7 years in automation and AI. 4.5 years freelance: 45 clients, 150+ workflows shipped, 99.9% uptime.

I build production pipelines in n8n, Python, and LangChain. CRM integrations, incident response systems, RAG pipelines, AI agents. Most clients come with a manual process that's eating engineering time — I replace it with something that runs unsupervised and fails loudly when it breaks.

- Groupon (2025-2026): 4-step incident response pipeline in n8n with LangChain at each triage point. Manual escalations down 80%. RAG documentation processor with Gemini: manual handling down 40%. HYPR: 28 Google Apps Script modules for 500+ employees' performance reviews.

- Brevo (2024): Lead generation cascade from LinkedIn through 6 enrichment sources with AI scoring. Processing time down 50%. Salesforce sync at 5,000+ records/day, 99.9% uptime for 18+ months.

- Freelance: 150+ workflows in n8n, Make, and Zapier for 45 clients. Lead enrichment via Cognism and Lusha: +25% conversion. AI agents for customer support, marketing automation, and logistics.

Every delivery includes error handling, centralized logging, and documentation readable by whoever touches it after me.

Stack: n8n, LangChain, Python, JavaScript, SQL/SOQL, Claude/Gemini/OpenAI APIs, Salesforce, HubSpot, PostgreSQL, BigQuery, REST APIs, Google Apps Script.
  • Francés

    Bilingüe o nativo

  • Inglés

    Competencia profesional completa

  • Español

    Competencia profesional completa

Acepta trabajo presencial
Madrid (hasta 50 km)

Experiencia

  • Groupon
    Automation & AI Developer
    julio de 2025 - abril de 2026 (9 meses)
    Madrid, España
    ▸ Incident Response Pipeline (n8n) Suite of 4 interconnected workflows (WF1–WF4) triggered when Jira tickets transition to "Mitigated". Auto-creates incident report docs, collects approvals, escalates blockers. Result: -80% manual chasing across the on-call team.

    ▸ AI Incident Reports Production RAG architecture using LangChain + Gemini, with PostgreSQL and BigQuery as sources. Generates structured incident reports from Jira, Google Chat and Google Docs. Standardized post-incident analysis. Result: -40% manual processing time on 200+ tickets/month.

    ▸ HYPR Performance Review System — Google Apps Script application (28 modules) using Gemini via LiteLLM to generate performance reviews for 500+ employees. Most complex project on the team. Owned spec, build, deployment, HR rollout.

    ▸ Jira ↔ Asana Sync (GIF — Global Initiative Framework) — Bidirectional framework for project approval flows. Approval bypass logic and cross-functional workflow unification.

    ▸ Kindly Reminder — Compliance monitoring system with configurable rules, automated weekly reports, and Jira auto-correction.

    Stack: n8n · Google Apps Script · JavaScript · Python · LangChain · Gemini · LiteLLM · PostgreSQL · BigQuery · Jira · Asana · Google Workspace APIs

    Process: discovery with stakeholders → spec with KPIs → 2-week sprints (POC → prod) → docs + on-call training → 30-day post-launch monitoring.

    Mission ended April 2026 currently exploring next senior AI/Automation role.

    | Key skills | n8n, Google Apps Script, JavaScript, Python, LangChain, Gemini, LiteLLM, PostgreSQL, BigQuery, Jira, Asana, Google Workspace APIs
    Langchain Atlassian JIRA n8n Python BigQuery
  • SaaS
    Automation & AI Consultant — Freelance (in parallel of employed roles)
    enero de 2021 - junio de 2025 (4 años y 5 meses)
    Madrid, Spain
    Designed and shipped 150+ workflows for 45+ clients across B2B SaaS scale-ups (martech, sales tools), e-commerce (DTC retail), marketing agencies, and tech startups. Maintained 99.9% uptime across the portfolio with centralized monitoring and Slack alerts on 60+ live workflows.
    Notable engagements:

    ▸ Plezi ↔ Pipedrive Bidirectional Sync Production-grade contact sync handling 10k+ contacts/month with retry logic, schema mapping and dedup. Result: 30% time saved on the client's Sales operations team.

    ▸ Lead Enrichment Pipeline Multi-source enrichment cascade combining Cognism and Lusha APIs, with quality scoring before CRM entry. Result: +25% conversion lift on the client's outbound campaigns.

    ▸ Enterprise AI Agents Built and deployed agents for marketing copywriting, 24/7 customer support, and logistics planning. Use cases ranged from FAQ deflection to content generation at scale.

    ▸ Centralized Workflow Monitoring Logging stack across 60+ live workflows with Slack alerts on failures, latency spikes, and SLA breaches. Reduced mean-time-to-detect from hours to minutes.

    Typical engagement: 1-4 week missions or monthly retainers.
    Owned scoping, technical design, build, client handoff and post-launch support.

    Stack: n8n · Make · Zapier · Python · JavaScript · OpenAI · Cognism · Lusha · Salesforce · HubSpot · Pipedrive · Brevo · Plezi · Slack

    Key skills | n8n, Make, Zapier, Python, JavaScript, Plezi, Pipedrive, Cognism, Lusha, OpenAI, Slack |
    LLMs Zapier n8n Make (Integromat) AI Agents
  • Brevo
    RevOps Engineer — API Workflows & Automation
    marzo de 2023 - abril de 2025 (2 años y 1 mes)
    Madrid, Spain
    ▸ Cascade Lead Generation Pipeline (n8n) End-to-end pipeline: LinkedIn scraping → multi-source verification (Cognism, Kaspr, Lusha) → AI validation scoring ICP fit and data quality. Cut lead processing time by 50% and improved data hygiene before CRM entry.

    ▸ Bidirectional Brevo ↔ Salesforce CRM Connector Production connector with custom JavaScript anti-duplication logic, sync of contacts, accounts and opportunities. Sustained 5,000+ records/day with 99.9% uptime in production. Owned design, code, monitoring, and incident response.

    ▸ Planhat Rollout Implemented Customer Success platform with native integrations to Salesforce, BigQuery, Harvstr, Jira and Zendesk. Delivered 360° customer view to CS team, consolidating data previously scattered across 5 tools.

    ▸ Two Zendesk Apps:
    (1) Real-time analytics dashboard surfacing ticket volume, response times, and SLA breaches by segment.
    (2) ChatGPT-powered assistant for content quality improvement directly inside agents' workflow.

    Cross-functional collaboration with Sales Ops, Marketing Ops, Customer Success, and engineering teams. Process: requirements gathering → spec with

    ROI estimate → build with monitoring → deployment + team training → maintenance.

    Stack: n8n · JavaScript · Python · Salesforce REST API · Brevo API · Planhat · Zendesk · ChatGPT · BigQuery · Harvstr · Jira

    Key skills | n8n, JavaScript, Salesforce, Brevo API, Planhat, Zendesk, ChatGPT, BigQuery |
    Zendesk n8n Make (Integromat) Salesforce Lightning JavaScript

Recomendaciones

Sé el primero en recomendar a Matthieu

Ayuda a este freelance a destacar compartiendo tu experiencia.

Estos perfiles de freelance también coinciden con tus criterios

AgathaA

Agatha Frydrych

Backend Java Software Engineer

4.7

(3)

2

BaptisteB

Baptiste Duhen

Fullstack developer

4.6

(4)

5

AmedA

Amed Hamou

Senior Lead Developer

4

(2)

7

AudreyA

Audrey Champion

Web developer

4.3

(3)

4

Formación

  • Full-Stack Web Development Bootcamp
    Le Wagon
    2021
    Part-time bootcamp covering full-stack web development: Ruby on Rails, JavaScript, HTML/CSS, SQL, Git, REST APIs, deployment. Capstone project shipped to production. Pivot point from marketing-tech background into engineering-grade automation work.
  • Master 2 (Programme Grande École) — Digital Marketing, Web Strategy & Artificial Intelligence
    Inseec Business School
    2019
    Master's program combining digital marketing strategy, web technologies, and applied AI. Focus on data-driven decision-making, marketing analytics, and emerging AI use cases in business operations. Foundation for later transition into RevOps Engineering and AI/Automation work.

Conjunto de habilidades profesionales

Categorías