About me
Ricardo Brossard
Systems engineer · AI and automation
I am a systems engineer (UTN) working on applied artificial intelligence, automation and integrations. I design, build and maintain end-to-end systems for businesses in any industry: from architecture and database to WhatsApp integration, deployment and maintenance.
At a glance
- Profession
- Systems engineer · AI and automation
- Education
- Systems Engineering, National Technological University (UTN), Argentina
- Location
- Argentina (UTC-3), remote work
- Languages
- Spanish; English: advanced technical reading and professional written communication
- My own product
- TomaNota
What I build
Artificial intelligence and agents
Assistants and agents that work inside your business: they answer customers on WhatsApp or on the web, read emails and documents, classify, reply and run tasks in your systems. With clear limits: data, numbers and money go through tested code, and anything that calls for judgment is decided by a person.
Integrations and automation
I make the tools you already use talk to each other, so information flows from one to the next on its own. If you already use n8n or Make, what you have keeps working. Whatever carries business rules moves to code in Python or TypeScript, with tests like any other software.
Custom apps and systems
Web platforms and apps for Android and iPhone, management systems and customer portals, with online payments and a screen for each person involved. Designed around how your team works.
Maintenance and evolution
Maintenance, improvements and new features on what we built. The system keeps running, adapts to changes in the business and grows in stages.
Projects
Systems I designed and built end to end. Each one has its full case study.
- Multi-merchant delivery, Argentina · In productionZizuA delivery platform for every shop in a city, with online payments.152end-to-end tests on every change, plus 57 unit testsRead the case study
- Real estate agency in Buenos Aires · In production, 10 usersCRM LexintonA real estate agency's inquiries, in one system with no manual entry.3,500+inquiries processed with no manual entryRead the case study
- My own product, for shops · In productionTomaNotaAn assistant that takes orders on WhatsApp without being able to invent a price.900+automated testsRead the case study
- For a real estate agency in Buenos Aires · Tested with real data, code publishedTasadorA property valuation report in one or two minutes, with every price justified.1 to 2 minper valuation report, against an hour or more of manual searchingRead the case study
Technologies
- Frontend
- ReactNext.jsTypeScriptViteTailwind CSSAstro
- Backend and data
- Node.jsFastifyPythonFastAPIPostgreSQLSupabaseRedis
- Artificial intelligence
- LangGraphRAGpgvectorTool callingVercel AI SDKLiteLLM
- Integrations
- WhatsApp Cloud APIMercado PagoGmail APIGoogle Calendar APIWebhooks
- Quality and deployment
- PlaywrightVitestpytestGitHub ActionsDockerVercelCloudflare
What I write
- Six channels, zero manual entry: how a real estate agency's inquiries come in on their own
- An AI that chooses instead of writing: from 3 seconds to half a second with Jev
- AI agents that don't make up numbers: where the model ends and the code begins
- The official WhatsApp API inside your system: a shared inbox and follow-ups that stop on their own
- Before you switch on the AI, measure it
Want to tell me what you need?
Message me on WhatsApp(opens in a new tab)