Prompt Engineer

This is a fully remote position, open to applicants in Brazil.

📋 Description

• Develop, test, and enhance prompts for text, image, audio, and multimodal models.

• Create straightforward, sequential, parameterized, and tool-oriented prompts.

• Construct frameworks for personas, context, few-shot prompting, structured reasoning, classification, extraction, generation, function calling, APIs, and information retrieval.

• Establish reusable libraries of prompts, templates, and AI components.

• Tailor prompts to various models, providers, languages, and application scenarios.

• Convert product and business needs into LLM-driven solutions.

• Engage in the development of agents, chatbots, copilots, and intelligent automation systems.

• Design conversational and decision-making workflows.

• Integrate AI models with APIs, databases, enterprise systems, and external tools.

• Collaborate with RAG architectures, tool-using agents, multi-step workflows, conversational memory, classification, and routing.

• Develop test suites and benchmarks for prompt evaluation.

• Set metrics for accuracy, relevance, consistency, completeness, hallucination rate, format compliance, latency, and execution cost.

• Assess models and prompting strategies; perform standard, edge-case, ambiguous, and adversarial testing.

• Implement ongoing evaluation in production; review failures and modify prompts, contexts, data, and validation criteria.

• Recognize and reduce risks associated with prompt injection, information leakage, inappropriate content, unauthorized responses, tool misuse, and data exposure.

• Draft safety instructions, refusal protocols, and operational limits.

• Adhere to Brazil’s LGPD, privacy guidelines, and AI governance practices.

• Establish methods for response validation, filtering, and auditing.

• Promote responsible AI usage policies.

• Organize documents, metadata, and knowledge bases for AI applications.

• Assist in document segmentation, indexing, and retrieval.

• Assess the quality, timeliness, and reliability of information sources.

• Design structured input and output formats, particularly JSON and JSON Schema.

• Collaborate with Product, Software Engineering, Data, Security, Legal, and business teams.

• Conduct interviews with users and subject-matter experts; document decisions, standards, versions, and experimental outcomes.

• Provide mentoring to engineers, analysts, and junior professionals.

• Effectively communicate technical limitations, risks, and findings to non-technical stakeholders.

• Monitor production models and detect performance variations.

• Manage versioning for prompts and configurations.

• Optimize cost, speed, stability, and scalability.

• Develop fallback, human review, and error-handling strategies.

• Stay informed on advancements in models, frameworks, and generative AI methodologies.

⛳️ Requirements

• Extensive experience with LLMs and multimodal models.

• Proficient in prompt engineering techniques.

• Familiarity with Python or a similar programming language.

• Experience with AI model APIs.

• Knowledge of RAG, embeddings, vector databases, and semantic search.

• Understanding of agents, function calling, and workflow orchestration.

• Comprehension of machine learning, natural language processing, and model evaluation.

• Experience with JSON Schema, REST APIs, databases, and distributed systems.

• Knowledge of observability, automated testing, and version control.

• Understanding of cloud computing.

• Familiarity with DevOps and MLOps is advantageous.

• Experience in industries such as Finance, Legal, Healthcare, Customer Service, or Technology is a plus.

• Involvement in AI projects that have been deployed to production is a plus.

• Experience with human evaluation and test dataset development is beneficial.

• Knowledge of fine-tuning, LoRA, and supervised training is a plus.

• Experience with LLM observability and evaluation platforms is a plus.

• Familiarity with AI application security is advantageous.

• Proficiency in advanced technical English is a plus.

• Demonstrated success in reducing costs, enhancing accuracy, or achieving measurable automation is a plus.

🏝️ Benefits

• Meal or food allowance.

• Discounts on courses, universities, and language institutions.

• Stefanini Academy — a platform offering free, current online courses with certificates.

• Mentoring opportunities.

• Benefits club providing discounts on consultations and medical exams.

• Medical insurance coverage.

• Dental insurance coverage.

• Benefits and discount club featuring offers at leading establishments.

• Travel club membership.

• Pet benefits program.