AI Workflow Engineer (LLM Integration & Prompt Engineering)

About OpenTrain

OpenTrain is a central job board for data-labeling and AI-training work. It aggregates open positions from many AI companies and labeling platforms so contributors can discover projects in one place and apply quickly.

Creating an OpenTrain account is free and applying typically takes only a few minutes.

• OpenTrain focuses on human-facing work that helps train and evaluate AI systems.
• This listing comes through the OpenTrain platform to connect skilled contributors with AI teams.

About AI Training Work

AI training (data labeling, annotation, or human feedback work) is the human side of building AI. People prepare, evaluate, and improve examples that models learn from—tasks include prompt design, response review, evaluation rating, fine-tuning, and workflow automation.

These projects are often fully remote and flexible: many roles allow you to choose hours and workload, and they range from entry-level annotation to specialist work requiring domain expertise.

• Contributors directly shape how models behave and improve accuracy, safety, and usefulness.
• Work types include prompt engineering, RLHF, fine-tuning, evaluation ratings, and API-based integrations.

The Role

Title: AI Workflow Engineer (LLM Integration & Prompt Engineering). This intermediate, contract, part-time role focuses on integrating large language models into automation workflows and improving model outputs through advanced prompting and iterative evaluation.

You will design real-world AI automation pipelines, refine prompts and input structures, create evaluation criteria, and review generated outputs for quality and reliability across tasks such as content generation, communication, and personalized outreach flows.

• Employment types: Contractor, Part-time.
• Time requirement: 20+ hours/week.
• Data type: Text. Labeling software: Other.

What You'll Do

Work hands-on with LLMs and automation tools to design, test, and refine prompts, prompt chains, and structured output formats. Iterate on prompts and inputs to improve accuracy, relevance, and safety.

Integrate LLM APIs into workflows using REST APIs, SDKs, or automation platforms; build and validate webhooks or automation steps that feed model inputs and handle outputs.

• Develop and test prompt templates, chaining strategies, and instruction designs for reliable outputs.
• Create and apply rubric-based evaluations to rate LLM outputs and identify failure modes.
• Implement input cleaning, templating, and data structuring to improve model performance in automation flows.
• Review model outputs for content generation, communication, and personalized outreach, documenting issues and improvements.

Requirements

You must meet the following experience and submission requirements exactly as stated in the project listing.

Submit a CV in English that indicates your level of English proficiency and includes an email address and phone number.

• Experience evaluating LLM outputs for legal reasoning quality.
• Hands-on text annotation, evaluation, or rubric-based QA experience.
• Experience integrating LLM APIs via REST APIs, SDKs, or automation tools.
• Strong prompt engineering background (instruction design, chaining, structured outputs).
• Skilled in building AI-powered content automation for communication.
• Proficient at data structuring, templating, and input cleaning.
• Knowledge of LLM limitations and mitigation methods.
• Experience with workflows using APIs/webhooks/automation tools (for example, Zapier).

Compensation, Labeling Details, and Level

Pay structure: PAY_PER_HOUR in USD. The listing shows an hourly rate of 45 (hourly range 15–45).

Labeling activities on this project include evaluation ratings, fine-tuning, RLHF, coding/function-calling related tasks, and other text-based annotations.

Experience level: Intermediate.

• Label types: EVALUATION_RATING, FINE_TUNING, RLHF, COMPUTER_PROGRAMMING_CODING, FUNCTION_CALLING.
• Labeling software: OTHER.

How To Apply and Location Restrictions

Apply via OpenTrain by submitting your CV in English that shows your English proficiency level and includes email and phone contact details. Only applicants who meet the listed requirements should apply.

This project is available worldwide except for the restricted locations listed below; please do not apply if you are in any of those places.

• Restricted locations for acquisition: Iran, Cuba, North Korea, Syria, Sudan, Venezuela, Myanmar, Russia, Belarus, Palestine.
• Restricted locations continued: Switzerland.
• Restricted locations continued: China, Taiwan.
• Restricted locations continued: Kenya.
• Restricted U.S. states: Alaska, Arkansas, California, Connecticut, Delaware, Georgia, Hawaii, Illinois, Indiana, Kansas, Louisiana, Maine, Maryland, Massachusetts, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, Ohio, Oregon, Tennessee, Utah, Vermont, Washington, West Virginia.
• Other restricted territories: Antarctica, Aruba, Åland Islands, Saint Barthélemy, Bonaire, Sint Eustatius and Saba, Bouvet Island, Cocos (Keeling) Islands, Democratic Republic of the Congo, Cook Islands, Christmas Island, Western Sahara, Falkland Islands (Malvinas), French Guiana, Guadeloupe, South