• Architect the inference pipeline that turns Life360 into a context layer LLMs can reason over: soul files, family context, behavioral signals, document understanding. • Choose models. Choose hosting. Decide what is self-hosted, API-based, fine-tuned, or prompt-engineered. Revisit those decisions as the landscape changes. • Build the serving infrastructure. Latency budgets, batching, caching, fallbacks, graceful degradation. Make it run at scale before scale gets here. • Build the eval loop. Know whether changes make the product better or worse, not just whether the demo works. This is the difference between a real product and a magic trick. • Own the cost model. Track spend per user, per feature, per family. Find the levers (context compression, model routing, pre-computation) that keep unit economics working from a test group to millions. • Define what gets persisted, what gets summarized, and what gets thrown away. The architecture decision of "what does our data look like to an LLM" lives with you. • Set the bar for safety, privacy, and trust at the infrastructure layer. Sensitive categories filtered before they propagate. Encryption, access controls, audit trails built in, not bolted on.

• Projetar, desenvolver e evoluir **prompts, chains e fluxos conversacionais** para aplicações baseadas em LLMs, garantindo consistência, precisão e alinhamento ao contexto do negócio; • Definir e implementar **estratégias de prompting** (zero-shot, few-shot, chain-of-thought, tool use), avaliando impactos em qualidade, custo e latência; • Criar e manter **mecanismos de avaliação de qualidade**, incluindo métricas automáticas, testes A/B e análises qualitativas, com uso de datasets de validação e benchmarks; • Estabelecer e aplicar **guardrails**, padrões de resposta e políticas de segurança, visando mitigação de alucinações, vieses, respostas inadequadas e comportamento fora do escopo esperado; • Colaborar na definição de **estratégias de segurança e privacidade**, prevenindo vazamento de dados sensíveis (PII), garantindo aderência às políticas internas e regulamentações; • Atuar na **melhoria contínua das aplicações de IA**, analisando logs, feedback de usuários e resultados de avaliação para refinamento dos modelos e fluxos; • Trabalhar de forma integrada com times de **produto, engenharia e UX**, apoiando a incorporação de soluções de IA em produtos digitais, APIs e experiências de chat; • Documentar decisões técnicas, padrões de uso e boas práticas, contribuindo para a escalabilidade e manutenção das soluções desenvolvidas.

• Design, implement and evolve advanced prompts, reasoning chains (chains), and interaction flows with LLMs to produce responses that are consistent, useful, and aligned with the business context; • Develop and maintain quality evaluations for generative AI solutions, including defining metrics, creating validation datasets, and continuously monitoring performance; • Define and apply guardrails, response patterns, and strategies to mitigate hallucinations, ensuring safety, reliability, and compliance with internal and regulatory policies; • Collaborate with product, engineering, and business teams to translate requirements into LLM-based solutions, supporting technical decision-making; • Support the implementation of security and privacy best practices, including prevention of sensitive data leakage (PII) and responsible use of language models; • Contribute to controlled experiments, such as A/B tests, prompt tuning, and comparison of approaches to continuously improve the user experience; • Document solutions, learnings, and technical patterns, promoting reproducibility, scalability, and knowledge sharing within the team.

• Design, develop, and optimize advanced prompts, reasoning chains (prompt chains), and multi-step flows for LLM-based solutions, ensuring consistency, accuracy, and scalability; • Define and implement quality evaluation strategies, including automated metrics, A/B testing, internal benchmarks, and continuous analysis of production results; • Establish safety guardrails, response standards, and control mechanisms to mitigate hallucinations, biases, and undesirable model behaviors; • Develop and maintain prompting standards (zero-shot, few-shot, system prompts, tool use), documenting best practices and promoting reuse across teams; • Contribute to the definition and evolution of LLM-based architectures, including RAG (when applicable), model selection, technical trade-offs, and alignment with business requirements; • Ensure compliance with security, privacy, and data protection policies, including prevention of PII leakage, response auditing, and responsible use of AI models; • Collaborate with product, engineering, and UX teams to integrate AI solutions into real products, ensuring good user experience, observability (logging, metrics), and reliability; • Analyze failures, user feedback, and production data to continuously iterate on prompts, evaluations, and solution architectures; • Stay up to date with trends and developments in the generative AI ecosystem, proposing technical improvements and innovations applicable to the business.