Senior AI Developer

This position is listed on behalf of a partner company, which manages all applications and next steps. Our partner is looking for a Senior AI Developer based in Brazil. As a Senior AI Developer, you will design, develop, and orchestrate advanced generative AI agents for voice and chat-based customer service applications. You will build intelligent conversational workflows that prioritize low latency, high accuracy, and reliable performance in production environments. Working with Python, large language models (LLMs), and modern AI orchestration frameworks, you will integrate AI capabilities with enterprise systems and legacy platforms. You will leverage Oracle Cloud Infrastructure (OCI) GPU resources and implement observability solutions to monitor performance, costs, and response quality. In close collaboration with Site Reliability Engineers and data scientists, you will help optimize deployments and continuously improve AI-generated responses. This role offers the opportunity to shape scalable, secure, and production-ready AI solutions within a collaborative, technology-driven environment.

Accountabilities

• Design, develop, and maintain Python-based orchestration services for generative AI agents supporting voice and chat channels.
• Build and optimize conversational workflows that deliver low-latency responses, high accuracy, and consistent user experiences.
• Implement and manage large language model orchestration using frameworks such as LangChain and/or LlamaIndex.
• Configure and maintain AI security and safety layers, including guardrails designed to improve response reliability and reduce inappropriate or unsafe outputs.
• Integrate enterprise tools and legacy systems, including CRM and billing platforms, through APIs and Model Context Protocol (MCP) servers.
• Implement comprehensive observability for AI agents using Langfuse, including trace instrumentation and monitoring of latency, performance, costs, and response quality.
• Collaborate with Site Reliability Engineering (SRE) teams to deploy, operate, and maintain AI applications on Oracle Cloud Infrastructure, including Oracle Kubernetes Engine (OKE).
• Work closely with data scientists to evaluate and refine LLM responses, improve conversational quality, and support continuous model and workflow optimization.
• Apply prompt engineering techniques to improve instruction following, contextual relevance, and the accuracy of AI-generated responses.
• Integrate enterprise services securely through REST APIs and mutual TLS (mTLS), following established security and communication standards.
• Troubleshoot production issues, identify performance bottlenecks, and implement improvements to the reliability, scalability, and efficiency of AI applications.
• Contribute to engineering best practices for deploying, monitoring, and maintaining production-grade generative AI solutions.

Requirements

• Advanced Python development skills, with experience building reliable, maintainable applications and backend services.
• Professional experience developing and orchestrating LLM-powered applications using LangChain, LlamaIndex, or comparable frameworks.
• Practical knowledge of Langfuse or similar observability platforms for monitoring AI agents, tracing execution flows, and analyzing performance and costs.
• Strong prompt engineering skills, including the ability to design, test, and refine prompts for complex conversational use cases.
• Experience integrating enterprise applications and services through REST APIs, with knowledge of secure communication mechanisms such as mutual TLS (mTLS).
• Understanding of generative AI architectures, conversational workflows, and the challenges involved in deploying LLM-powered applications in production.
• Ability to diagnose technical issues, optimize application performance, and improve the reliability and accuracy of AI-driven solutions.
• Strong collaboration and communication skills, with the ability to work effectively alongside SREs, data scientists, and multidisciplinary engineering teams.
• A proactive, analytical mindset and commitment to continuous improvement, technical quality, and secure software development practices.

Nice To Have

• Familiarity with NVIDIA NeMo Guardrails or similar frameworks for implementing safety and security controls in generative AI applications.
• Knowledge of the Model Context Protocol (MCP) and experience building or integrating MCP servers.
• Experience with th…