AI/ML Software Engineer — Benchmark Task Designer

Obsidian; San Francisco (CA); On-site; USD 68,880 - 103,320; Part time

Job summary

Obsidian is looking for a professional to author complex tasks aimed at evaluating AI models on professional document understanding and instruction following. The role requires a minimum commitment of 15–20 hours per week and ideally seeks candidates with over 3 years of experience in software engineering or data science.

Responsibilities include crafting tasks that assess an AI’s ability to reason over technical documents, with a focus on accuracy and structure. This position offers remote work opportunities.

Qualifications

• Minimum commitment of 15–20 hours per week.
• 3+ years of hands-on experience in software engineering or data science.
• Experience with AI models and technical documentation is a plus.

Responsibilities

• Author tasks to evaluate AI models on document understanding.
• Define clear evaluation criteria and outputs.

Skills

Software engineering

Data science & analytics

Job description

Obsidian is looking for a professional to author complex tasks aimed at evaluating AI models on professional document understanding and instruction following. The role requires a minimum commitment of 15–20 hours per week and ideally seeks candidates with over 3 years of experience in software engineering or data science.

Responsibilities include crafting tasks that assess an AI’s ability to reason over technical documents, with a focus on accuracy and structure. This position offers remote work opportunities.