Product Engineer · São Paulo, Brazil

I build the products
I used to sell.

Engineer with an unusual path. I started as a developer 20 years ago, spent five years on the business side of Brazilian B2B SaaS startups (Head of Growth, CRO), and went back to engineering in 2023. Today I'm at Tailor, a B2B SaaS within ZAX (backed by Atlantico, FJ Labs, GFC), where I ship real solutions to users every day and lead product engineering end-to-end.

Italian (EU) citizen · São Paulo (UTC−3) EN · PT · IT
Currently shipping at Tailor: AI agents on WhatsApp Business API · 20+ ERP integrations · Stripe at scale.
100+
B2B brands
using Tailor
2k+
Daily users
on the platform
20+
External integrations
shipped
2.5k+
Commits in the
last year

Things I have built.

Where I am actually useful.

I am not the engineer you hire to optimize your sorting algorithm. I am the one for messy real-world problems where the hard part is the decisions, not the code. APIs that misbehave, customers who change their minds, ERPs from 2003, payments that need to settle in three currencies, AI features that have to actually work in production.

i

Integrations that hold

Stripe webhooks, WhatsApp Business API, OpenAI and Anthropic Claude APIs, ERP integration layers. I build connectors that do not break the moment a vendor changes a header.

b

Backend that scales linearly

Node.js, Python, PostgreSQL on AWS and Kubernetes. Pragmatic schema design, idempotent webhooks, async queue-based flows. Not flashy. Just survives production for years.

r

Revenue-aware product calls

Five years running revenue means I can read a P&L and a Stripe dashboard at the same time. Useful when the next feature decision is also a pricing decision.

What I actually use.

No framework worship, no resume-driven decisions. The stack below is what is running in production today across Tailor, ContaLaudo, and the integration layer in between.

Languages

  • TypeScript
  • JavaScript
  • Python
  • SQL

Frontend

  • React + Next.js
  • TypeScript
  • Tailwind CSS
  • Server-side rendering

Backend & APIs

  • Node.js + Express
  • Python (FastAPI, Flask)
  • REST APIs
  • GraphQL
  • Async / queue-based pipelines

Data & Cloud

  • PostgreSQL
  • Prisma / SQLAlchemy
  • Supabase
  • AWS
  • Kubernetes + Docker

Integrations & AI

  • Stripe (subscriptions, webhooks)
  • WhatsApp Business API
  • OpenAI API
  • Anthropic Claude API
  • AI agents in production

What people I have worked with say.

André has an incredible learning capacity — I was always impressed by how fast he absorbs and synthesizes new information. That ability directly contributed to weme's growth, with strong impact on the growth and product teams he led. Beyond being smart, the team trusts him and looks up to him for advice. He was always the person to help me think beyond the obvious, on any topic.

CK
Carolina Kia Takada

Former CEO — weme · MIT Under 35

We went from a 100% manual order process to a fully integrated digital operation. The result was immediate: 120 orders processed in a single day, with zero human error.

BF
Bruno Felipe

Commercial Director — Apparel, Fortaleza

We rolled out AI-driven lead qualification and unlocked a critical bottleneck — 100% of leads are now triaged in real time without depending on the team.

DA
David Albuquerque

Founder — B2B Wholesale, São Paulo

The unusual path.

I started as a developer 20 years ago. Then I deliberately crossed into business — five years building revenue engines and leading commercial teams in Brazilian B2B SaaS. The detour was not a pivot; it was a way to learn what engineers do not see: how money moves, why margins die, what the customer actually buys. In 2023 I returned to engineering with that lens permanently installed.

2024 — now

Product Engineer & Director — Tailor / ZAX

Building B2B SaaS end-to-end. Architecture, integrations, AI features, revenue tooling. About 80% of my time in code, the rest on product calls, hiring, and team.

2023 — 2024

Founder & Product Engineer — Métoddo

Built tech solutions for the fashion and apparel industry in Brazil. Started as growth consultancy, pivoted into building software for clients. The transition back to engineering started here.

2021 — 2023

Chief Revenue Officer — Vesti

Built and led the Revenue function — marketing, sales, customer success, payments. Hands-on with the tooling underneath: internal automations, BI dashboards, Hubspot integrations. The role that taught me where margins die and why "ship faster" is sometimes the wrong answer.

2017 — 2021

Growth & Revenue Leadership

Head of Product and Head of Growth roles across Brazilian B2B SaaS startups (weme, Sensedia, Digiage, Magnética). Mentor at Gama Academy. The years that wired product, growth, and revenue thinking together.

2010 — 2016

Professional Online Poker Player

Six years of professional cash-game play. Decision-making under uncertainty, expected-value reasoning, financial discipline, pattern recognition with incomplete information. The mental models I still run on every day as an engineer.

2005 — 2009

Software Developer & Electrical Engineering

Where it all started. Building web applications and studying Electrical Engineering (UFMA, then Universidade Anhanguera São Paulo).

Things I think about.

What does "Product Engineer" actually mean in practice?
Owning a feature from the product call to the deploy. Not just the code, but the trade-off about what to build, the integration choices, the pricing implication, and the data model that won't haunt the team in 18 months. The label exists because it captures that scope.
How has AI changed what you build day to day?
It changes what's worth building. Features that used to require six months of model work and a data team can now ship in two weeks with OpenAI or Claude — but only if you treat them as production systems, not demos. That means observability, retry logic, prompt versioning, evals, fallbacks. The interesting work is on the boring parts.
What makes a B2B SaaS integration actually hold up in production?
Idempotent webhooks, dead letter queues, version pinning where the vendor allows it, and the discipline of treating every external API as eventually consistent. The hardest part isn't the happy path — it's the day a vendor pushes a header change and your queue starts dying.
Why fashion-tech specifically?
A $445B industry in Brazil alone with an obvious tooling gap — except it isn't obvious, because the operational reality is full of contradictions: suppliers who quote in WhatsApp, returns that never settle to zero, ERPs from 2003 that refuse to die. The kind of messy world worth building software for.
What's the hardest part of going from revenue back to engineering?
Calibrating my own intuition. Five years on the business side made me reach for shortcuts that weren't there — "we'll fix it in implementation", "the team will figure it out". Engineering punishes that. The reset took longer than I expected, and I still catch myself.

Always up for a good conversation.

Always up for a 20-minute chat about B2B SaaS in emerging markets, AI agents in production, building integration layers that hold, or the engineering–revenue overlap. Drop me a line — I read everything.