Remote AI Research Engineer: Agentic Post-Training

There is Talent; Ciudad de Mendoza; A distancia; ARS 278.728.000 - 371.638.000; Jornada completa

Descripción de la vacante

Tether is hiring an AI Research Engineer to advance post‑training methodologies for agentic systems and tool use. You will refine pre‑trained models to reason, plan, and autonomously invoke tools, including on edge devices, while handling multimodal inputs.

The role targets state‑of‑the‑art performance through data pipelines, distributed training, and rigorous evaluation, collaborating with cross‑functional teams to production‑readiness and real‑world impact.

Formación

• Requires degree in CS/ML or related field; advanced degree preferred.
• Proven experience with multimodal post-training workflows and data pipelines.
• Demonstrated open‑source contributions on platforms like GitHub or Hugging Face.

Responsabilidades

• Conduct end‑to‑end research and engineering for post‑training of agentic and tool‑use models.
• Drive cross‑cutting model improvements: factuality, instruction adherence, tool usage, multi‑agent coordination.
• Design and enhance large‑scale post‑training systems, data pipelines, and benchmarking infra.
• Develop rigorous evaluation suites and diagnostic tools for model readiness.
• Strengthen feedback loops from real‑world product usage; incorporate signals into post‑training.
• Collaborate with tooling, product, and training teams to improve usefulness of frontier models.

Conocimientos

Multimodal post-training workflows

Distributed training on multi-node GPU

Reasoning, tool use, multi-agent

Open-source contributions (GitHub/Hug"

Publications at AI conferences

Educación

Degree in Computer Science / ML or related field

MS/PhD preferred

Herramientas

GitHub

Hugging Face

Descripción del empleo

Tether is hiring an AI Research Engineer to advance post‑training methodologies for agentic systems and tool use. You will refine pre‑trained models to reason, plan, and autonomously invoke tools, including on edge devices, while handling multimodal inputs.

The role targets state‑of‑the‑art performance through data pipelines, distributed training, and rigorous evaluation, collaborating with cross‑functional teams to production‑readiness and real‑world impact.