Senior AI Engineer - Internal Solutions

Role overview This role focuses on building the next generation of an agentic customer support system for a Latin American instant b2b payments platform. The engineer will own the support automation layer end-to-end, replacing ad hoc workflow tools with a unified decision engine that serves multiple channels. The mission is to deliver a reliable, secure, multi-tenant AI system that resolves client questions in real time and escalates gracefully.

Responsibilities - Own the agentic support layer from migration off ad hoc workflow automation onto a single decision engine, extending across portal chat, WhatsApp, and Slack entry points. - Build deterministic scaffolding around the nondeterministic core, including event-driven pipelines, idempotency, retries, dead-letter queues, typed domain models, and audit trails. - Enforce strict client-data isolation as a first-class architectural constraint, preventing any cross-tenant data exposure. - Integrate against documentation, public APIs, sandbox environments, data warehouse, and CRM systems behind clean, well-tested interfaces. - Design confidence-based escalation paths that hand off cleanly to Customer Success or Integrations specialists with full context. - Instrument the system with correlation IDs, resolution-rate dashboards, time-to-first-response metrics, and meaningful alerting. - Maintain model and vendor choices behind an anti-corruption layer for swappable providers and frameworks.

Requirements - 5+ years building and operating production backend services with post-launch ownership. - Depth in at least one backend-focused language, with willingness to work primarily in Go alongside Python. - Distributed systems fundamentals including asynchronous processing, at-least-once delivery, idempotency, and eventual consistency. - Cloud-native AWS experience with Lambda, DynamoDB, S3, SQS/SNS, containers, infrastructure as code, and CI/CD. - Architectural judgment using hexagonal architecture and domain-driven design with rich models and ports-and-adapters discipline. - Testing discipline with high coverage as a gate, table-driven and contract tests, and honest approaches for nondeterministic components. - Experience with multi-tenant sensitive data systems and isolation discipline. - Working knowledge of LLM-backed production features including prompt design, tool-calling, retrieval, and fallback strategies. - Professional fluency in Spanish (working language) and working English for technical documentation.

Nice to have - Experience with agent-orchestration frameworks and evaluating multi-step, tool-calling agents. - Experience building or maintaining MCP servers or comparable tool-exposure layers. - Familiarity with golden datasets, eval suites, and confidence thresholds for AI workflows. - Kafka, schema registries, and contract evolution experience. - Experience migrating products off workflow-automation tools onto centralized services. - Payments, treasury, or fintech background, especially client integrations or support tooling.