João Gonçalves
15 years scaling engineering teams and DevOps. Shipped AI into product within a month of ChatGPT's launch. Now I run 14 AI agents that ship full-stack features end-to-end.
Founding Engineer at BRIDGE IN, running a 14-agent AI orchestration system that ships full-stack features end-to-end. 15 years of engineering leadership before that — DevOps, platform, and engineering departments through one $20M acquisition and several scale-ups.
Engineering used to be where everything stalled. Now it’s everywhere else: product clarity, decision speed, the ability to evaluate output that nobody on the team wrote.
The teams pulling ahead aren’t using AI more carefully. They’re using it harder, with their hands on the wheel. Senior judgment plus AI velocity. Depth in the system, ruthless review of output, willingness to throw the model’s first answer away.
I’ve spent 15 years in the old pattern: scaling teams, owning DevOps and platform functions, leading a department through a $20M acquisition. I started shipping AI into product within a month of ChatGPT’s launch, before the playbook existed. I now run the new pattern at BRIDGE IN, with agents handling most of the code. The leverage is different. The work is the same.
Founding Engineer at BRIDGE IN. Building the BRIDGE IN Operating System, a Django/React platform that centralizes payroll, accounting, HR, legal, and tax compliance for international companies across four European markets.
The unusual part is how it gets built. A 14-agent AI orchestration system runs a 4-phase workflow (plan, implement, test, deliver) through 27 custom development skills I authored. Sentry triage, Slack feedback processing, autonomous issue creation, CI failure resolution, production deployment. The agents operate as a permanent member of the engineering team. I direct the system. The system does most of the work.
The receipts: in a recent month the system merged 30 agent-authored PRs at 97% autonomy, 29 of 30 with no human commits, every one through the full CI gate. It triaged Sentry alerts into filed-and-fixed issues autonomously, caught regressions before users did, and turned reported feedback into shipped fixes at a 4.9-hour median. Median PR lead time, open to merge, was 7 minutes.
The hard part isn’t agents writing code. It’s keeping them reliable across parallel git worktrees, where permission and isolation failures are the real operational hazard. That’s the actual work: state coordination, quality gates, and recovery loops that let the system run without a human babysitting each step.
The product itself has AI in the workflow, parsing documents, classifying inputs, surfacing the right action before a human touches it. AI in how we build, and AI in what gets built.
- Building the BRIDGE IN Operating System from the ground up. A full-stack platform (Django, React) centralizing payroll, accounting, HR, legal, and tax compliance for international companies across four European markets.
- Designing and operating a 14-agent AI orchestration system where specialized agents coordinate through a 4-phase workflow to plan, implement, test, and deliver full-stack features end-to-end with minimal human intervention.
- Authoring 27 custom development skills that automate the entire engineering pipeline: from Sentry error triage and Slack feedback processing to autonomous issue creation, CI failure resolution, implementation, and production deployment.
- Operating AI agents that autonomously monitor production errors, triage user feedback, implement high-priority fixes, and generate release reports, functioning as a permanent member of the engineering team.
- Establishing the entire product development function from zero — CI/CD, quality gates, error monitoring, and release cadence in under 3 months.
- Operating the system at 97% agent autonomy in a recent month (29 of 30 agent-authored PRs merged with no human commits), with autonomous Sentry triage and a 4.9-hour median from reported feedback to shipped fix.
- Integrating Valispace as an Altium 365 app and ensuring continuity and merging capabilities in the aftermath of a high-profile acquisition.
- Directing the team that built a production RAG assistant inside the Valispace requirements platform: retrieval over engineering requirements, test data, component trees, and company-specific authoring standards, surfaced to roughly 200 enterprise customers, the largest with over 1GB of requirements text. GPT in the cloud, Llama for on-prem. I owned the architecture and the technical calls.
- Doubling the engineering department's size and coaching new team leads to own all operations, encompassing project management, system architecture, and software development.
- Interfacing cross-departmentally to identify and implement technology solutions that directly impact business growth, streamline processes, and optimize data analytics capabilities, notably equipping DevOps with improved Kubernetes infrastructure to accommodate a surge in demand.
- Cultivating a healthy, productive, and meritocratic department culture that retained 90% of its staff post-merger.
- Positioned Valispace, a fast-growing startup, as an attractive asset for a $20M acquisition by Altium, and directed all due diligence to ensure a frictionless transaction and transition to new ownership.
- Partnered with the CEO & CPO to orient the company's strategic direction and pioneered AI's early adoption.
- Orchestrated an engineering department's 25+ staff on three teams to successfully implement cloud computing and other emerging technologies, notably operating a complex global cloud environment with only three engineers.
- Implemented ISO-27001 to access untapped and highly regulated opportunities in the aerospace sector, such as Airbus, Clearspace, and iSpace.
- Oversaw an operations team of four responsible for 100+ cloud-based & on-premise deployments with maintenance & support.
- Owned a robust continuous integration and deployment pipeline that accelerated time-to-market for software releases by 88.89%.
- Established and maintained strong relationships with external vendors and stakeholders to ensure the successful integration of third-party tools and services.
- Performed in-depth code reviews, amplified team productivity, reduced software defects, and streamlined development processes by introducing Agile and automated testing frameworks.
- Developed, maintained, and improved both the backend REST API and the Frontend app.
- Built and continuously improved a low-code tool that was ahead of its time because it enabled non-technical users to assemble complex software solutions like ERPs, HR portals, and document management resources; clients included Portugal's government & armed forces.
- Conceived an IoT modular platform, built a working prototype, and pitched the project to a larger tech company to secure financing and merge capabilities.
- Led the Partnering Place Mobile app: project management, UI render engine designed around descriptive models, Phonegap-based delivery with Backbone.js, Angular.js, and per-OS UI customization.
- Built and continuously improved a low-code tool that enabled non-technical users to assemble complex software solutions like ERPs, HR portals, and document management resources; clients included Portugal's government & armed forces.
- Built remote management applications across mobile devices and web admin tools, on Phonegap with JavaScript frameworks.
- Web applications (HTML, CSS, JavaScript, Python, Django, SQL, AJAX, ActionScript) for intranets, newsletters, multimedia, and database systems.
Not actively looking, but I take a few conversations a year that are interesting enough to make space for.
The fits that map best: forward-deployed or applied AI engineering, agent and AI-infrastructure work, AI solutions architecture, or founding / early-stage technical leadership, at a company building at the edge of agents and engineering or shipping AI into a real product. Also open to advisory work with operators figuring out how to put AI into their delivery loop without producing more slop. Remote from Lisbon, open to travel.
What I won’t: anything pitched as “AI transformation,” roles where AI is a buzzword, companies looking for a hire to validate a strategy already decided.
If that fits, the email and LinkedIn below work.