SergioGongora

I connect the partsthat make software think.

Next.js, Supabase, and the APIs that make products actually work.

Full-stack engineer · 7+ years

Scroll

What I actually do

Four things, done properly.

Systems integration

Payments, comms, data, and AI arrive as separate systems. The work is making them behave like one.

5systems shipped end-to-end

Real-time data

Webhook-driven sync with idempotent writes and retry-safe ordering — current state, not last night's export.

realtimenot batch

Multi-agent AI

Scoped agents with defined jobs and defined handoffs, coordinated and observable instead of one oversized prompt.

6agents coordinated in production

End-to-end delivery

Schema to deploy, including the parts nobody finds interesting. Localization and payouts are architecture, not afterthoughts.

7+years shipping product

Every product is a pile of disconnected systems —payments, comms, data, AI models — until someone wires theminto something that behaves like one coherent organism.

The stack

As it actually connects.

Every tool below is one I've shipped with. Select one to see which projects actually use it — the mapping comes from the case studies, not from a wishlist.

Frontend

Infrastructure & data

AI

Payments & comms

Data visualization

Select a tool above to trace it through the work.

Shipped with, in production

Next.jsReactTypeScriptTailwind CSSVercelSupabasePostgreSQLStripe
ResendWorkvivoAI ModelsAgent OrchestrationD3.jsLive ChartingThis Site

Timeline

The real sequence.

  1. Foundations

    Full-stack fundamentals

    Years spent building the muscle that now powers every project below: shipping product end-to-end, not just the parts that are interesting.

  2. Marketplace & payments

    Two-sided marketplace infrastructure

    Designed and built the transaction backbone for a physical-ad marketplace — matching, bidding, and payouts wired together as one system.

  3. Real-time sync

    Cross-platform data sync

    Built the sync layer keeping a third-party engagement platform and a product's own membership data in lockstep, in real time.

  4. Multi-agent AI

    Production AI agent orchestration

    Took AI systems past the demo stage — six coordinated agents running in a real product, each with a defined job and a defined handoff.

  5. Bilingual delivery

    Multi-market product delivery

    Shipped a bilingual (ES/EN) product end-to-end, treating localization as a first-class architectural concern, not an afterthought.

  6. Now

    Agentic mobile product

    Current focus: an agentic coding app on mobile — AI systems that act, not just answer.

Get in touch

Establish a connection.

Reach out about a role, a project, or a system that needs wiring together.