Engenheiro(a) de IA Sênior — Python e Agentes (LangGraph)

Accountabilities:: Maintain and evolve AI agents: Develop, maintain, and improve existing agentic systems built with LangChain and LangGraph, contributing to both architectural decisions and implementation. Build business-specific agent tools: Design and implement tools that enable AI agents to execute real operational tasks for healthcare clinics, integrating these capabilities with internal APIs and external services. Develop WhatsApp integrations: Connect AI agents to WhatsApp-based communication channels, supporting automated interactions with patients and clinic teams through reliable messaging workflows. Integrate large language models: Implement production-ready LLM capabilities, including tool calling, structured outputs, and retrieval-augmented generation (RAG), to support accurate and context-aware agent behavior. Ensure workflow reliability: Design robust execution flows with persistent state across conversations, asynchronous processing, queues, retries, and appropriate error handling. Monitor agent performance: Measure and optimize response quality, latency, and operational costs using evaluations, structured logs, and performance metrics. Implement testing and quality standards: Develop automated tests and maintain sound software engineering practices to ensure reliability, maintainability, and predictable behavior in production. Strengthen application security: Identify and address risks associated with LLM-powered applications, including prompt injection, sensitive data exposure, and privacy requirements. Collaborate with product teams: Work closely with Product to plan technical deliverables, assess feasibility, prioritize improvements, and translate business needs into effective AI solutions. Provide technical leadership: Serve as a technical reference for AI engineering, share knowledge, support architectural decisions, and mentor colleagues when needed. Drive continuous improvement: Identify technical bottlenecks, propose pragmatic solutions, and balance delivery speed with code quality and long-term system reliability. Requirements: Python development: Strong experience developing and maintaining production applications in Python, particularly with FastAPI and Pydantic. AI agent frameworks: Practical experience with LangChain and LangGraph, or another agent framework, combined with a willingness to work extensively with LangGraph. Production LLM integration: Demonstrated experience integrating large language models into real applications, including tool calling, structured outputs, and retrieval-augmented generation (RAG). Database expertise: Experience working with relational databases such as PostgreSQL or MySQL. Software engineering fundamentals: Knowledge of automated testing with Pytest, Clean Architecture, and best practices for building maintainable, testable, and scalable applications. Asynchronous processing: Experience implementing queues, webhooks, messaging systems, or other asynchronous processing mechanisms. Problem-solving and ownership: A proactive mindset, strong accountability, and the ability to investigate and resolve problems beyond the immediate scope of assigned coding tasks. Collaboration and communication: Ability to work effectively with product and engineering colleagues, communicate technical decisions clearly, and contribute to shared goals. Learning and adaptability: Humility to teach, learn, ask for help, and adapt to evolving technologies and practices in the rapidly changing AI landscape. Pragmatic decision-making: Ability to balance speed, quality, reliability, and maintainability when delivering AI-powered solutions. LLM evaluation (preferred): Experience evaluating AI outputs and using observability or evaluation tools such as Langfuse or LangSmith. Messaging integrations (preferred): Familiarity with the WhatsApp Cloud API or official Business Solution Providers (BSPs). Cloud and infrastructure (preferred): Experience with Docker and AWS services, particularly ECS/Fargate, SQS, or Amazon Bedrock. AI security and privacy (preferred): Knowledge of prompt injection prevention, sensitive data protection, and compliance with Brazil's General Data Protection Law (LGPD). Technical leadership (preferred): Previous experience mentoring developers, guiding technical decisions, or serving as a technical lead. Healthcare technology (preferred): Familiarity with healthcare software, clinical workflows, or applications designed for medical practices. Benefits: Competitive compensation: Market-aligned remuneration under a service provider (contractor) arrangement. Fully remote work: Work from home with no regular office attendance required. Birthday leave: A dedicated day off to celebrate your birthday. Well-being benefits: Access to Wellhub for fitness and wellness activities. Paid annual break: Annual paid recess to support rest and work-life balance. Collaborative work environment: A relaxed, supportive atmosphere that encourages teamwork, knowledge sharing, and continuous learning. Professional growth: The opportunity to take ownership of AI engineering initiatives, expand your expertise in agentic systems, and contribute to the evolution of production AI applications. Meaningful impact: Help develop intelligent automation solutions that improve daily operations for healthcare professionals and their patients. Working schedule: Availability during standard business hours, Monday through Friday.