Remote AI Research Engineer: Agentic Post-Training

YO AI Labs; Ciudad de Mendoza; A distancia; ARS 230.602.000 - 353.590.000; Jornada completa

Descripción de la vacante

Tether invites you to join a globally distributed AI research team focused on agentic post-training, tool invocation, and multi-modal models. You will refine pretrained systems to reason, plan, and autonomously call tools for edge devices and real-world tasks.

You will work on scalable post-training pipelines, evaluation suites, and robust deployment workflows, contributing to state-of-the-art capabilities and edge-case resilience.

Formación

• PhD/MS in Computer Science or ML with publications in AI conferences.
• Experience in post-training for agentic workflows and tool use.
• Proven open-source contributions related to agentic systems.

Responsabilidades

• Advance post-training of agentic models and tool usage.
• Design data pipelines and evaluation workflows.
• Collaborate with cross-functional teams to deploy updates.
• Improve model safety, reproducibility and efficiency.

Conocimientos

LLM architectures

Agentic workflows

Tool use

Reinforcement learning

Multimodal data

Open-source work

Educación

MS/PhD in CS/ML

Herramientas

Distributed training frameworks

Multi-node GPU

GitHub/Hugging Face

Descripción del empleo

Tether invites you to join a globally distributed AI research team focused on agentic post-training, tool invocation, and multi-modal models. You will refine pretrained systems to reason, plan, and autonomously call tools for edge devices and real-world tasks.

You will work on scalable post-training pipelines, evaluation suites, and robust deployment workflows, contributing to state-of-the-art capabilities and edge-case resilience.