AI Platform Engineer — Agentic Systems & Infrastructure

Location: Remote — Latin America

Employment Type: Full-time

Compensation: USD $8,000–$11,000/month

Working Hours: Meaningful overlap with US Pacific Time

Company Overview

A fast-growing, seed-stage AI company building self-learning and real-time adaptation infrastructure for AI agents. The company works with leading frontier AI labs and large technology organizations and is scaling its engineering team to meet rapidly increasing demand.

About the Role

We are looking for an AI Platform Engineer who combines strong platform and infrastructure engineering experience with hands-on ownership of production agentic systems.

You will work directly with the founders and senior engineers to build the core platform powering reinforcement learning environment delivery. This is a high-ownership role for someone comfortable working across backend systems, cloud infrastructure, agent architectures, evaluations, and occasional frontend needs.

This is not a RAG-only or chatbot integration role. We are looking for engineers who have shipped agentic systems into production and handled real-world failures and edge cases.

What You’ll Do

• Build and maintain backend services and cloud infrastructure using Python, AWS, Docker, and Kubernetes
• Design and ship production-grade agentic systems, including agent harnesses, sandboxes, multi-agent architectures, and evaluation frameworks
• Identify failure modes, create evaluations, and continuously improve agent behavior
• Own technical decisions and systems from initial design through production
• Contribute to TypeScript and React applications when needed
• Participate in architectural decisions and help shape the future engineering team
• Collaborate daily with the US-based founding team
• Join the founding team for an in-person onboarding week in the US during the first month

Must-Haves

• 3–9 years of professional software engineering experience
• Experience shipping agentic AI systems end-to-end in production
• Hands-on experience with agent harnesses, sandboxes, multi-agent systems, or agent evaluation frameworks
• Strong Python, AWS, Docker, and Kubernetes experience
• Solid backend and platform system design skills
• Ability to investigate real-world edge cases and improve systems with limited data or tooling
• Strong ownership and product sense, with the ability to influence technical decisions and roadmap
• Experience working in fast-paced, high-intensity startup environments
• Professional English communication skills
• Based in Latin America and able to overlap meaningfully with Pacific Time

Nice-to-Haves

• TypeScript and React fundamentals
• Experience with reinforcement learning environments or AI research tooling
• Previous experience at an early-stage or zero-to-one startup
• Data science or machine learning background that transitioned into production agentic system

Interview Process

1. Application review
2. Introductory call
3. Take-home exercise
4. Technical discussion
5. Work trial
6. Offer