Remote AI/ML Engineer: Agentic Posttraining & Tool Use

Fullstack; Tandil; Remote; ARS 185,448,000 - 278,173,000; Full time

Job summary

Tether is seeking an AI Research Engineer focused on agentic posttraining to advance how models reason, plan, and autonomously invoke tools across edge devices. You will refine pre-trained models to enhance intelligence and tool use in real-world tasks.

The role involves developing and testing post-training algorithms for goal-directed behavior, working on multi-modal systems, and contributing to the optimization of tool invocation in a fully remote, worldwide team.

Qualifications

• Deep expertise in large language model architectures.
• Extensive post-training experience for agentic workflows.
• Experience with tool use fine tuning and function calling.
• Reinforcement learning from feedback on multi-turn interactions.

Responsibilities

• Curate agentic training data (tool use, reasoning chains, environment interactions).
• Strengthen baseline performance and robustness of agentic systems.
• Identify and resolve bottlenecks in post-training for tool augmented agents.
• Develop and test new post-training algorithms to enable goal-directed behavior.

Job description

Tether is seeking an AI Research Engineer focused on agentic posttraining to advance how models reason, plan, and autonomously invoke tools across edge devices. You will refine pre-trained models to enhance intelligence and tool use in real-world tasks.

The role involves developing and testing post-training algorithms for goal-directed behavior, working on multi-modal systems, and contributing to the optimization of tool invocation in a fully remote, worldwide team.