Mid-level AI Python Developer (Backend)

CI&TAI software

Listing location: Brazil; Workplace type: Remote; Structured job location: BrazilFull-TimeMiddle

Salary not disclosed

Apply NowOpens the employer's application page

Job Details

Use LLMs and prompt engineering techniques, including hallucination mitigation, context-window management, and structured output generation with JSON or Pydantic.
Build and maintain RAG pipelines, including ingestion, chunking, embedding generation, and searches in vector databases.
Connect LLMs securely to external systems using function calling (tools), with strong input and output validation.
Have solid backend development experience with Python.
Create and consume REST APIs, webhooks, and authentication.
Debug and analyze model responses, and help create test cases to ensure prompt or data-source changes do not break existing flows.
Use AI orchestration frameworks such as LangChain, LangGraph, LlamaIndex, or similar.
Use LLM tracing tools such as Langfuse or LangSmith to debug calls and analyze latency.
Use Git, Docker, and relational and NoSQL databases.
Have basic knowledge of queues and asynchronous processing, such as Redis, SQS, or Celery.
Familiarity with automated evaluation methodologies such as LLM-as-a-judge is an advantage.
Interest in and knowledge of Model Context Protocol (MCP) is an advantage.

Develop and maintain standardized APIs and webhooks for AI agent tools, with security and rigorous error handling.
Support the design and maintenance of conversational state machines and graphs to manage context across long conversations.
Monitor and adjust knowledge-base context retrieval, refining the data provided to LLMs.
Work with senior engineers to discuss architectural improvements.
Analyze production metrics and implement security and orchestration layers.

View Full Description & ApplyYou'll be redirected to the employer's site