Software Developer & AI Trainer — JSON, Tables, Native Languages
Use your software development skills to design and edit JSON, tables, and prompts in one native language to train generative AI models. Remote contractor role at $25/hr with a minimum 15 hours/week (preferred 20+), must be able to start Aug 19.
Generative AI & RLHF
100% Remote Hourly · $25/hr
$25/hr
Compensation
Worldwide
Eligibility
Intermediate
Experience
Aug 14, 2024
Posted
Open worldwide
Interested in this role?
Create a free OpenTrain account and apply in minutes.
OpenTrain is the #1 platform for people building careers in AI training and data labeling. We hire and contract contributors to create, review, and improve the examples that teach modern AI systems how to understand code, text, and structured data.
Creating an OpenTrain account is free — the platform makes it easy to find projects, build a profile, and apply in minutes so you can start impactful, flexible work quickly.
Why AI training work matters
AI training (also called data labeling or human feedback work) is the human side of building intelligent systems: people prepare examples and prompts that models learn from. This is a fast-growing, remote-friendly industry with flexible schedules and meaningful influence on how state-of-the-art systems behave.
100% remote — work from anywhere with a computer and internet connection.
Flexible, part-time-friendly schedules — fit work around studies, another job, or family.
Accessible roles — many projects need language fluency and attention to detail rather than prior AI experience.
The role — what this project is
You will create, edit, and validate structured data formats (JSON, tables, lists) and design prompts in one native language to train and evaluate generative AI models in software development and engineering contexts.
This is a contract, part-time contributor role. OpenTrain is hiring developers who can produce precise, well-structured training examples that reflect realistic programming workflows and instructions.
Work type: Contractor, part-time.
Data type: Text (structured JSON, tables, lists, prompt text).
Labeling tasks: prompt design, content generation, accuracy and relevance checks in a specified native language.
Key responsibilities
You will be responsible for producing high-quality structured data and prompts, checking format correctness, and ensuring examples are accurate and contextually appropriate for software engineering tasks.
Create and edit JSON files, tables, and lists that represent prompts, inputs, and expected outputs.
Write clear, natural prompts and responses in your native language for software-focused scenarios.
Validate structure, syntax, and completeness of JSON and tabular data.
Ensure content aligns with provided guidelines and software development context.
Requirements — must have
Candidates must be native speakers of exactly one of the specified languages and meet the software development and language proficiency requirements below. Please confirm your native language when you apply.
Native speaker of ONE of these: Korean (South Korea), Japanese (Japan), German (Germany), Italian (Italy), Spanish (Spain or Mexico only), Chinese (Simplified, Hong Kong, or Taiwan), Portuguese (Brazil), or French (France).
Proficiency in English at B2 level or higher.
Strong background in software development: 2+ years professional experience OR 4+ years formal software development education.
Hands-on experience with structured data formats (JSON, tables, lists).
Proficiency with programming languages such as Python, JavaScript, SQL, and HTML/CSS.
Excellent attention to detail and ability to produce precise structured data and prompts.
Able to start the project next week (August 19) and commit a minimum of 15 hours/week (preferred 20+ hours/week).
Preferred experience
These are not strict requirements but will make your application stronger and improve your qualification score.
Experience writing data and prompts for LLM training or other generative-AI projects on AI-training platforms is a plus.
2+ years of creative or technical writing experience in your native language.
Proven ability to create and manipulate JSON prompts for model training and evaluation.
Strong problem-solving and critical-thinking skills applied to developer-focused tasks.
Schedule, pay, and employment details
This is a paid, hourly contractor role. OpenTrain hires contributors worldwide, but you must be a native speaker of one of the specified languages as listed above.
Pay: USD $25 per hour (hourly contractor).
Time commitment: minimum 15 hours/week; project expects 20+ hours/week where possible.
Employment types: Contractor, Part-time.
Work format: Fully remote, text-based annotation and prompt design.
How to apply and selection notes
Apply through your OpenTrain account. In your submission, state your native language and summarize your software development background, years of experience, and examples of JSON or prompt work you've done.
Applications will be evaluated for language fluency, developer background, and ability to produce clear JSON and prompt examples. Candidates with prior LLM prompt-writing or AI-training writing experience will be scored higher.
When applying, explicitly state your native language and whether your Spanish or Chinese background matches the allowed dialect/country requirements.
Include links or short samples showing JSON or prompt examples where possible.
You may be asked to complete a brief qualification task demonstrating prompt design and JSON formatting.
Contract JavaScript/TypeScript developer to build and evaluate AI models and datasets (20+ hrs/week, remote worldwide). Combine full-stack development with model benchmarking, response ranking, RLHF/SFT support, and written evaluation rationales.
Review and improve AI-generated frontend code by writing reference implementations, explaining reasoning, and ranking model responses. Expert HTML/CSS, accessibility, and cross-browser knowledge required; 20+ hrs/week, $70/hr, hiring across multiple countries.
Join OpenTrain AI to train and evaluate AI systems on API integration scenarios, working remotely 20+ hours/week with pay from $15–$45/hr. You will design prompts, review LLM-generated integration plans and payloads, and test robustness across REST APIs, webhooks, and SaaS connectors.