Swift iOS Developer, LLM Training
Key details
- Work type
- remote
- Employment
- contract
Job Description
About OpenTrain AIOpenTrain AI is the #1 platform for people building careers in AI training and data labeling. We hire and contract contributors who create the examples and feedback that teach modern AI systems how to behave.
This role is hired directly by OpenTrain AI and offers remote, flexible work on projects that shape how real-world developer tools and assistants perform.
About AI training workAI training (data labeling/annotation) is the human side of building artificial intelligence: people prepare and review examples—code, text, audio, and images—that models learn from.
It’s an accessible, rapidly growing way to work in tech, often part-time and fully remote.
Contributors help set model quality and safety by writing prompts, producing model-style responses, and grading outputs so models can solve real developer problems in production languages like Swift.
The roleWe are looking for Swift developers to craft detailed coding prompts and high-quality responses to train LLMs to write, explain, and debug Swift/iOS code. You will work with AI researchers and dataset teams to build examples that reflect real-world programming challenges.
This is a contractor, part-time role requiring 20+ hours per week. Compensation is hourly at $12 USD.
Job type: Contractor, Part-timeTime commitment: 20+ hours/weekPay: $12 USD per hourData type: Text; Label type: Computer programming / codingWhat you'll doDaily tasks focus on producing and reviewing code-centric training data so models learn to generate, refactor, and explain Swift code correctly and idiomatically.
Write clear, instructional prompts that pose realistic Swift/iOS tasks (implementation, debugging, refactors, explanations).Create model-style responses showing correct code, step-by-step reasoning, and concise explanations.Review and rate model outputs for correctness, style, and safety; provide corrective examples when needed.Collaborate with AI researchers to refine dataset guidelines and edge-case coverage.RequirementsCandidates must demonstrate practical Swift and iOS experience and strong English communication.
All requirements below are essential or explicitly noted as preferred.
Bachelor’s or Master’s in Engineering, Computer Science, or a related field, or equivalent experience.Minimum 2 years of professional experience in software development; ideally 2–3+ years focused on iOS/Swift.Highly proficient in Swift; experience with SwiftUI and Objective-C is a plus.Proven experience building and publishing commercial-grade iPhone and/or iPad apps.Strong command of written English for writing prompts, responses, and feedback.Previous experience creating training data, labeling, or working with LLMs is highly preferred.Who should applyThis role is a great fit for iOS engineers who enjoy teaching and explaining code, care about API design and best practices, and want flexible, remote, part-time work that directly shapes developer-focused AI.
You do not need prior annotation experience to apply if you meet the Swift and iOS requirements, but LLM/data-labeling experience is a strong advantage.
How applications and work proceedApply with examples of your Swift work (links to apps or code snippets) and a short note about any prior experience producing training data for AI.
During onboarding you will complete small calibration tasks to confirm Swift skill and written-English communication. Work is performed remotely.
As a contractor you will be paid hourly at the listed rate and will receive task instructions and dataset guidelines from OpenTrain AI.
Be prepared to demonstrate Swift/iOS experience and to confirm English proficiency during screening.If you have prior LLM training or annotation platform experience, describe it in your application—this is a strong plus.
Trust-check this listing
Verify scam risk and ghost-job signals before you apply.
Related roles
Browse more remote Software Engineer jobs, or every remote job category.
Source: Google Jobs • Last updated 1w ago