Software Development Engineer, AI Agents

- Experience (non-internship) in professional software development
- Experience contributing to the architecture and design (architecture, design patterns, reliability and scaling) of new and current systems
- Experience programming with at least one software programming language
- Experience with full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations
- Bachelor's degree in computer science or equivalent
- Experience with Machine Learning and Large Language Model fundamentals, including architecture, training/inference lifecycles, and optimization of model execution

Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.
Amazon is hiring a Software Dev Engineer to focus on Agentic AI in Brazil - Sao Paulo, building the capabilities that let AI agents act reliably on real operator workflows. You'll join a specialist team of Product Engineers / Software Developer Engineers, Scientists, and Developers within Amazon's emerging-markets technology organization, focused on sustainable growth across Latin America and other emerging markets.

Your customers are operators. Planners, ops managers, analysts, support leads. They make hundreds of small judgement calls a day, and your job is to give them agent capabilities that absorb the repetitive ones so they can focus on the calls that actually matter.

In this role you are part builder, part product owner. You scope the pain, design the mechanism, build the prototype, instrument the metrics, then iterate in production. You build the primitives, integrations, and mechanisms behind agentic systems, integrating large language models with internal data and trading off latency, cost, and accuracy with judgement, not by recipe.

Key job responsibilities
- Review agent runs in production, triage operator corrections, and determine which are bugs versus signal that the autonomy boundary needs to move.
- Pair with operators on workflows you are scoping. Watch them work, ask why behind every step, and translate implicit rules into mechanisms the agent can execute.
- Close the feedback gap by ensuring every operator correction becomes a test, an eval, or a prompt update, not lost noise.
- Own the roadmap end to end. Prioritise, demo weekly, and articulate design choices in writing.
- Partner with operations leaders in each market to identify which workflows are agent-ready and which still need a human in the loop.
- Translate ambiguous operator pain points into concrete, measurable mechanisms with clear success criteria.
- Monitor trust score, accuracy, latency, and cost per decision. Dig into regressions before they compound.