[Job-32336] Mid-level IA Python Developer (Backend), Brazil

Brazil | Oct. 22, 2026

Report as Closed

Company: CI&T

Country: Brazil

Type: Onsite

Employment: Homeoffice

Description: At CI&T, we help large companies transform the potential of AI into real business impact with AI Deployment, AI-native execution and tech-integrated business solutions. With 30 years of experience in technological transformation, we accelerate innovation with expertise in agentic SDLC, application modernization, Data & AI, martech and business strategy.  We are 8,000 CI&Ters in more than 25 countries, collaborating to build solutions with real impact. AI is already part of the way we work, evolve and innovate every day. Important: if you live in the Metropolitan Region of Campinas, your presence in the city's offices will be mandatory, in accordance with the current attendance policy. CI&T is looking for a Full Python AI Engineer to join the AI ​​service agent development team, focusing on B2C and B2B journeys. Your role will be to integrate AI models with the real world, implementing tools via APIs, maintaining conversational flows and evolving RAG pipelines to ensure efficient and quality service. Responsibilities: Develop and maintain standardized APIs and Webhooks that will serve as the "hands" of the AI agent (Tools), ensuring security and rigorous error handling. Support the design and maintenance of state machines/conversation graphs, helping to control the context of interactions so that the AI ​​does not get lost in long conversations. Monitor and adjust context retrieval from the knowledge base, refining the data delivered to LLM to improve the accuracy of responses. Work together with Senior Engineers to discuss architectural improvements, analyze production metrics and implement new layers of security and orchestration. Requirements for this challenge: Practical understanding of using LLMs and Prompt Engineering techniques, including hallucination mitigation, context window management and generating structured output (JSON/Pydantic). Practical experience in building and maintaining RAG pipelines (ingestion, chunking, generation of embeddings and searches in vector databases). Ability to connect LLMs to external systems securely via Function Calling (Tools), with strong validation of input and output data. Solid experience in backend development with Python, consumption and creation of REST APIs, webhooks and authentication. Ability to debug and analyze model responses, supporting the creation of test cases to ensure that prompt or database changes do not break existing flows. Practical experience with AI orchestration frameworks such as LangChain, LangGraph, LlamaIndex or similar. Practical use of LLM tracing tools to debug calls and analyze latency (e.g. Langfuse, LangSmith or similar). Mastery of Git, Docker and relational/NoSQL databases. Basic knowledge of queues and asynchronous processing (Redis, SQS or Celery). [Optional] Differentials: Familiarity with automated EVAL methodologies (LLM-as-a-judge). Interest and knowledge in Model Context Protocol (MCP).     #LI-DJ1 #LI-midsenior

Apply here:

Web: Apply here

Emails: