Senior AI Engineer
Posted 4hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
Engenheiro de IA sênior construindo a plataforma de MLOps, LLMOps, RAG e agentes da GOK. Garantindo segurança, observabilidade e implantações controladas em ambiente financeiro regulado.
Responsibilities:
- Build and maintain pipelines with version control, automated testing, controlled deployment, and rollback.
- Implement MLOps, including model registry, training and inference pipelines, retraining, monitoring, and drift detection.
- Implement LLMOps, including a unified model gateway, prompt versioning, automated evaluation, and caching.
- Develop RAG and agent-based applications, implementing their harnesses, context management, skills, tools, and execution controls.
- Integrate solutions with data sources, APIs, and transactional systems, including integrations via MCP.
- Monitor quality, availability, latency, throughput, and costs, with usage-level traceability.
- Implement guardrails, access controls, and data protection measures in accordance with regulated-environment requirements.
- Investigate failures and maintain recovery procedures and operational documentation.
Requirements:
- Experience in software engineering applied to data and artificial intelligence systems.
- Proficiency in Python and experience with Spark.
- Experience implementing MLOps and LLMOps.
- Practical knowledge of agent harnesses, including execution lifecycle, state management, checkpoints, resumption, and failure handling.
- Knowledge of context engineering, including information selection, memory, token limits, and isolation between sessions and users.
- Experience with skills, tool calling, and MCP to provide agents with capabilities and tools.
- Knowledge of RAG, embeddings, retrieval, and context-relevance evaluation.
- Experience with cloud platforms, containers, CI/CD, and infrastructure as code.
- Strong understanding of the operational impact of latency, throughput, cost per request, and failure modes.
- Knowledge of AI security, including guardrails, permissions, secrets management, and protection against prompt injection.
- Experience in regulated environments with audit and traceability requirements.
- Willingness to work under a contractor (PJ) arrangement.

















