Prompt Engineer / AI Agent Developer

• Configure system prompts, instructions, and agent conversation flows on the client’s platform
• Compare and select the most suitable model for each agent, considering quality, cost, and latency
• Adapt instructions to the specific characteristics of each LLM
• Define response-quality criteria and conduct iterative testing with structured test cases
• Apply guardrails, handle out-of-scope responses, and implement mitigation strategies for prompt injection
• Document prompt engineering standards, versions, and best practices for reuse and team enablement

• Proven hands-on experience in prompt engineering on real-world projects using multiple LLM families, such as OpenAI/GPT, Anthropic/Claude, Google/Gemini, and Meta/Llama
• Hands-on experience developing AI agents in squads, including orchestration, tool use, and context-memory management
• Experience with metrics and quality testing for AI-generated responses
• Exceptional command of written Portuguese, with clarity, precision, grammatical accuracy, and appropriate tone
• Familiarity with Git and deployment through existing CI/CD pipelines
• Ability to quickly learn and operate new proprietary platforms
• Experience with RAG architectures, particularly evaluating responses based on documentary sources
• Knowledge of Python for test automation and LLM evaluation
• Experience working on projects involving sensitive data under Brazil’s LGPD or in the healthcare sector

• Meal or food allowance
• Discounts on courses, universities, and language institutions
• Stefanini Academy — a platform offering free, up-to-date online courses with certificates
• Mentoring
• Benefits club offering discounts on consultations and medical tests
• Health insurance
• Dental insurance
• Benefits and discounts club with offers at leading establishments
• Travel club
• Pet benefits program