Last Updated on August 13, 2026 by Justin Bryant
Cohere is seeking a remote Data Annotation Specialist in Canada for part-time work evaluating and improving artificial-intelligence systems. The listing gives a contract rate of CAD $30 per hour and an expected schedule of 16 hours per week.
This is an independent-contractor opportunity, not an employee position, and it is not described as entry-level. Cohere asks for at least one year of related experience. The company also warns that projects can involve explicit, offensive, or otherwise inappropriate text, and that the availability of continuous work is not guaranteed.
The exact job was present in Cohere's official Ashby job-board feed and listed as active when checked on August 13, 2026. It is limited to applicants in Canada.
What Is Cohere?
Cohere develops enterprise artificial-intelligence models and tools. Its technology helps organizations use large language models for tasks such as search, generation, retrieval, and workplace automation while maintaining control over business data and deployment.
Data-annotation and evaluation specialists help improve these systems by judging model responses, applying consistent standards, and creating reliable examples. The work is more analytical than the title may initially suggest.
Pay and Benefits
The posting lists a contract rate of CAD $30 per hour for approximately 16 hours per week. That would equal CAD $480 for a fully available 16-hour week before taxes and other contractor expenses, but the listing says projects and hours can fluctuate and does not guarantee continuous work.
This is an independent-contractor arrangement. Cohere explicitly says contractors do not receive employee health benefits. Contractors should also expect to handle their own taxes and should review the agreement carefully for payment timing, project assignment, confidentiality, and termination terms.
The role uses a bring-your-own-device setup, so the selected contractor needs an appropriate personal laptop. The posting does not advertise an equipment allowance.
Job Highlight: Data Annotation Specialist, Generalist
The specialist reviews the output produced by AI models and decides how well it follows instructions, uses evidence, stays accurate, and meets quality standards. Depending on the project, the contractor may rank multiple responses, correct text or structured data, write prompts and ideal examples, or apply detailed rubrics and taxonomies.
Although the title includes data annotation, this is not passive data entry. It requires judgment, careful reading, consistent documentation, and the ability to explain why one model response is better than another. Some tasks may use text, images, or structured formats such as JSON, CSV, TSV, Markdown, XML, or YAML.
The work itself appears largely computer-based, but it should not be labeled no-phone. Cohere describes a hiring process that includes a video screen, and remote teamwork may require meetings or other synchronous communication.
Key Responsibilities
- Evaluate AI-generated responses for accuracy, usefulness, safety, and instruction-following.
- Rank multiple model outputs and document the reasoning behind quality decisions.
- Stress-test models by designing challenging prompts or edge cases.
- Create datasets, prompts, reference answers, and strong examples for model training.
- Build or apply rubrics, taxonomies, and project-specific annotation guidelines.
- Label and correct text, image, and structured-data content.
- Participate in calibration exercises, so reviewers apply standards consistently.
- Identify patterns in model performance and report quality problems clearly.
- Meet project deadlines while maintaining careful, repeatable judgments.
Qualifications
Cohere asks for:
- At least one year of experience in AI data annotation, large-language-model evaluation, content moderation, research, or related analytical work involving quality assurance or preference ranking.
- Excellent written English and close attention to meaning, logic, and detail.
- Experience with annotation platforms or comparable review tools.
- Comfort working with structured formats such as JSON, CSV, TSV, Markdown, XML, or YAML.
- The ability to work independently, follow detailed instructions, and deliver consistent results remotely.
- A personal laptop that can be used for the work.
The hiring process may include a resume and writing-sample review, a virtual annotation test, a video screen, and an independent-contractor agreement. Applicants should not assume that general typing or office data-entry experience alone satisfies the one-year analytical-experience requirement.
Resume Keywords
If they truthfully match your experience, potentially relevant terms include:
- AI data annotation
- LLM evaluation
- Preference ranking
- Model-output evaluation
- Quality assurance
- Content moderation
- Prompt creation
- Dataset development
- Rubric design
- Taxonomy application
- Calibration
- Error analysis
- Structured data
- JSON, CSV, and TSV
- Markdown, XML, and YAML
- Written English
- Independent remote work
Use keywords only when you can support them with real examples. If you have worked under confidentiality, safety, or sensitive-content rules, describe that experience accurately without disclosing protected information.
Location and Schedule
- Remote eligibility: Canada only
- Workplace type: Remote
- Employment type: Independent contractor
- Schedule: Approximately 16 hours per week
- Rate: CAD $30 per hour
- Equipment: Bring your own laptop
- Work availability: Project-based; continuous work and fixed hours are not guaranteed
- Potential content exposure: May include explicit, offensive, or inappropriate text
- Ashby job ID: a7321948-288c-4e9f-9fbe-2f9ccda07ffa
- Published date in official feed: July 24, 2026
The listing does not establish a fixed daily shift or promise a long-term volume of hours. Applicants should confirm time-zone expectations, project duration, invoicing procedures, and whether any security software or device specifications apply.
Closing Notes
This opportunity may appeal to someone in Canada who already has AI-evaluation, annotation, moderation, research, or analytical quality-review experience and wants a limited part-time schedule. The transparent hourly rate and 16-hour target make the basic commitment easy to understand.
The major limitations are just as important: it is contractor work without employee health benefits, uses the contractor's own laptop, may expose reviewers to disturbing content, and does not guarantee continuous projects or steady hours. It is neither a no-experience opening nor simple data entry.
The official Ashby feed was last checked on August 13, 2026. Confirm that job ID a7321948-288c-4e9f-9fbe-2f9ccda07ffa remains listed and read the full terms before applying.
Note: If the “Apply Here” link no longer works, the job may have been filled. Subscribe below to be notified of future remote jobs.