About X402
The engineering firm that answers the 3 a.m. page.
We design, build, and operate the systems other engineers hand off. Production infrastructure, industrial automation, fintech rails, zero-trust networks — built by the people whose number is on the call when it matters.
Built by an engineer, not a sales team.
Mathias Cordeiro
I started X402 because the market had a gap: firms that could build the prototype, and firms that could manage the helpdesk — but almost no one who would design the system, write the firmware, commission the hardware, and then stay on to operate it when production broke at 3 a.m.
Before founding X402, I spent over a decade running enterprise infrastructure for companies that couldn't afford downtime. That included operating Brazil's leading startup and venture-capital hub (three buildings in São Paulo, ~1,000 active users a day — redundant firewalls, switches, PoE access points, cameras and access control, end to end) and the US operations of a South Korean financial group for 13 years (New York to San Francisco, three floors, 50+ professionals, 100+ managed devices). Physical operations are systems too, and that discipline shapes every technical decision we make.
My distributed systems background runs from zero-trust mesh networks (6,000+ endpoints, three datacenters on two continents) to fintech rails (stablecoin on-/off-ramp for three banks, 35,000+ end users) to industrial IoT at scale (4,073 machines, 21 factories, 3 countries, pilot live in 40 days). I have written firmware, laid out opto-isolated signal-conditioning boards, built autonomous drone agents, and co-built payment-rail platforms as part of core engineering teams.
X402 is deliberately small and senior. The engineer who scopes your project builds it. That's not a pitch — it's the only way we've found to deliver systems that actually survive contact with production.
Senior specialists. No generalists on critical paths.
X402 operates as a tight network of senior engineers, each with production-depth in their domain. We staff the discipline the project needs — not the headcount the contract allows.
Networks that defend themselves.
Zero-trust mesh architecture, enterprise firewalls, redundant datacenters, active-active replication, and AI-driven security layers. We don't hand off to a managed provider — we are the managed provider.
Hardware and firmware, in-house.
Opto-isolated signal conditioning designed by our own electrical engineers, PoE-powered single-board agents, custom firmware, and real-time OEE computation at the edge — from bare metal to the cloud dashboard.
Rails that settle in 3.4 seconds.
Payment-rail platforms, stablecoin on-/off-ramp, on-chain address provisioning per end user, and legacy-rail bridges. We've co-built core infrastructure for fintechs serving 35,000+ bank customers across three institutions.
Agents that fly, watch, and act.
Autonomous flight agents on 500+ drones, RTK-GPS boustrophedon sweep planning, AI security layers that inspect every mesh flow in real time, and LLM-augmented operations tooling — inference where it makes a measurable difference.
The custom operating environment.
When six fragmented business data sources need to become one coherent operating environment, we build it. ERP integration, real-time dashboards, multi-tenant data isolation, and two-way sync — designed to be operated by us long-term.
The building that enforces itself.
RFID and biometric access control wired into the ERP, captive portals that enforce billing at the network layer, lighting, climate and camera surveillance on open-source automation cores — hardened to enterprise grade.
Four commitments we don't negotiate on.
These aren't values-page platitudes. They're the constraints we impose on ourselves because every system on our portfolio runs in production and we're the ones who get paged when it doesn't.
Senior depth, every role.
No junior engineers on critical paths. Every discipline on a project is staffed by someone with production scars in that exact domain. We staff narrow and deep, not broad and thin.
We operate what we build.
Handing off a system to a third party after delivery is an incentive problem. When we run the production environment, we design for operability from the first commit — because the 3 a.m. page comes to us.
Open source where it wins. Proprietary where it doesn't.
We're not ideological about either camp. We've run every class of firewall, network OS, and platform in the trade — and we pick the best tool for each position. Vendor lock-in is a risk we manage, not a shortcut we take.
Redundancy is the default, not the upgrade.
Redundant firewalls, active-active replication, DR sites on separate continents, LUKS encryption at rest on every server — these aren't upsells. They're what you get when the people building the system are also the ones responsible for its uptime.
instrumented
rails we built
operations for clients
across 3 countries
Work with us.
We take on a small number of projects at a time so we can give each one genuine engineering attention. If you have a system that needs to work — and stay working — tell us what it does and what's breaking. You'll get a straight answer within a day, from the engineer who would build it.