Founding Engineer (Core Full-Stack) @ Mainder · Automation / Integrations / AI Systems
📍 Barcelona, Spain
I build production software across backend, frontend, automation, and infrastructure.
Most of my work sits at the intersection of integrations, worker systems, automation, and LLM-enabled workflows.
Frontend
- Next.js · React · TypeScript
- Tailwind · shadcn/ui
Backend
- Ruby on Rails · Node.js (ecosystem)
- Worker systems, scheduled execution, and job orchestration
Databases
- PostgreSQL · MySQL · MongoDB
Infra / Ops
- AWS (Flow & Ops)
- Railway
- Docker
Jobs / Queues
- Sidekiq
- Redis (queues, coordination, rate limiting patterns when needed)
Automation / Scraping
- Python · Playwright
- Browserless orchestration
- Robust automations (timeouts, retries, UI variance, popups/modals)
APIs / Integrations
- Third-party API integrations
- Webhooks, API contracts, authentication flows, and reliability patterns
LLMs / AI
- OpenAI · Gemini · Perplexity (API-driven workflows)
- Prompt-based automation patterns and structured outputs
AI Coding / Dev Tools
- Cursor · Codex · Claude Code
- MCP-based integrations to automate coding + operational workflows
I’m a founding engineer at Mainder (HRTech / Recruiting SaaS). We built the product from the start and we’ve owned critical systems across core backend, automation, and frontend:
- We designed and implemented the multiposting system as a Ruby on Rails microservice
- We built execution + orchestration around Redis + Sidekiq (scheduled/background workloads)
- We worked on database workflows, reliability patterns, and operational robustness
- We built and maintained backend workflows connecting product features, integrations, and data integrity
- We implemented production-grade worker flows and third-party API integrations
- I’m currently focused on the frontend with Next.js + TypeScript
- I’m reworking product flows with a modular, maintainable architecture
- I rebuilt the InfoJobs Worker: an autonomous worker that scrapes InfoJobs and performs actions inside the portal
- Implemented in Python + Playwright, designed to run end-to-end without manual intervention
- Built to handle layout changes, dynamic UI behavior, and real-world variance
I collaborate with TRAQEER on automation systems for detection + enforcement:
- Automated DMCA form submissions end-to-end (Playwright-based flows)
- Built pirated content discovery via search queries using ScrapingDog
- Integrated with the public Reddit API to find/report content and support takedown workflows
- Redesigning the onboarding flow with React
I built an AI conversational bot for an e-commerce context:
- Conversation flows for discovery, support, and product guidance
- Backend integrations to connect the assistant with catalog and business logic
- Email: emilianorozas@gmail.com
- Website: https://emiliano-rozas.com/
- LinkedIn: https://www.linkedin.com/in/emilianorozas/
- GitHub: https://github.com/HxrmxStudio



