Co-Founder & Lead Software Engineer
Founded and lead technical development of AI-powered livestock management platform using computer vision for cattle weight estimation. Scaled to 193 farms, 448 users, and 70,000+ animals with 93% accuracy.
Key Achievements
- Co-founded AgTech startup winning national 'Desafío AgTech 2023' competition
- Architected end-to-end cattle weighing automation system with type-safe TypeScript across frontend (Next.js) and backend (NestJS REST APIs), and dedicated Python/FastAPI microservices for ML inference
- Engineered scalable multi-tenant SaaS handling hundreds of concurrent users with sub-300ms API response times, async queue processing for ML workloads, and graceful error handling, achieving 99.8% uptime through monitoring with Sentry
- Designed hybrid architecture separating business logic (modular monolith) from ML processing (dedicated microservices), enabling independent scaling and deployment of compute-intensive operations while maintaining system reliability
- Led data-driven technical pivots: migrated from real-time to offline processing after production analysis showed reliability issues, and introduced dual-camera setup increasing model accuracy from 67% to 93% based on edge case analysis
- Reduced weighing time from 4 hours to 30 minutes per session, saving 50 hours/month per farm through process automation
- Deployed on AWS (EC2, S3, Lambda, SQS, RDS) and CI/CD through GitHub Actions
Responsibilities
- Own complete technical strategy from requirements to architecture and deployment
- Manage 3-person engineering team while remaining deeply hands-on with code across full stack, balancing strategic technical decisions with daily execution and feature development
- Drive product roadmap based on direct customer feedback from producers
- Design and maintain multi-tenant architecture supporting 193 simultaneous farms with 448 users
- Build offline-first synchronization for unreliable rural connectivity