דרושים AI Task Evaluator בכל הארץ

Apply here 👉 AI Agent Evaluation Analyst Application Form

[email protected]

We’re Hiring: AI Task Evaluator – TAU Framework (Remote | Contract)

We’re collaborating with a leading global AI company on the TAU (Tool–Agent–User) framework, an advanced benchmark designed to evaluate how AI agents perform in realistic, multi-step environments.

As part of this project, you’ll help test and refine AI reasoning by reviewing simulated real-world interactions — where an AI Agent uses tools to complete user requests while following business rules, policies, and logic.

🧠 Role Overview

You’ll analyze and annotate AI-agent conversations and task trajectories to determine if:

• The agent’s reasoning and tool usage are logical and consistent.
• The policies (privacy, accuracy, authorization) are respected.
• The final outcome matches the correct “golden path.”
• The conversation flow is realistic, clear, and aligned with the user’s intent.

This role blends quality assurance, research, and logic-based evaluation — ideal for people who love breaking down processes and improving system intelligence.

🔍 Key Responsibilities

• Review agent–user interactions and identify logical gaps or policy violations.
• Validate tool sequences and end results against golden sets.
• Flag inconsistencies, missing steps, or unrealistic actions.
• Annotate errors and reasoning issues clearly and concisely.
• Suggest edge cases or task improvements to enhance coverage and realism.

✅ What We’re Looking For

• Excellent analytical and critical-thinking skills.
• Strong attention to detail and logical consistency.
• Ability to understand structured workflows (JSON/YAML reading familiarity is a plus).
• Clear English communication and documentation skills.
• Background in QA, consulting, linguistics, research, or systems analysis preferred.