ufactor.ai is a lean team of senior engineers and architects, amplified by the same AI tools and agents we build for clients. We use AI agents to build agents — which is how a small team ships production-grade systems with senior hands on every part.
The team is led by Rui Brás Fernandes — 25+ years of technology leadership across multi-national companies including Cisco and Deloitte, two startups built and exited, and Agentic AI shipped to production.
Around that sits a deliberately small group of senior people. We stay lean on purpose and let AI carry the repetitive work — so the hands on your system are senior ones, not a junior team learning on your budget.
We spent years running the Software Development Lifecycle as u-factor.io — shipping products, exits, and enterprise platforms. The move to ufactor.ai isn't a rebrand; it's a new loop: the Agent Development Lifecycle. Building agents is 90% software, 10% AI — and the 90% is exactly what we already knew how to do.
Two acquired SaaS spinoffs. Renault & Citroën launches. Production systems for healthcare, last-mile, and events — the boring, hard parts of shipping software, at scale.
Multi-agent systems on Google ADK, LangGraph, and CrewAI. State, tools, evals, observability — the SDLC muscle made the pivot quick; the ADLC loop keeps it learning.
Real-time event logistics — hardware, firmware, and platform — built end to end and exited to a Renault-aligned operator.
Indoor positioning platform acquired by Crowdkeep, now part of Veea. The whole stack shipped to production.
A multi-agent platform for telco and enterprise operations — agents in production with human-in-the-loop and custom tooling.
A last-mile delivery platform powered by an AI Operations Engine — planning routes, resolving exceptions, and coordinating dispatch and drivers in real time.
In his Deloitte role, Rui acted as technical leader from the Deloitte side on a publicly reported agentic AI initiative with NOS for telecom network automation.
The initiative moved from architecture design to production agents in around four months, covering priority network operations workflows such as fault detection, incident management and Root Cause Analysis.
According to ECO, some RCA processes achieved above 80% efficiency / time savings, and the model is being explored for replication across international telecom operators.
Selected work. Names and dates shown for context — happy to walk through any of these in detail on a call.