At IX Coach, we believe in unlocking human potential through deep, authentic connection and transformative experiences. Our mission is to create AI tools that empower people to rediscover themselves, connect meaningfully with others, and live more fulfilling lives. This isn’t just about training models—it’s about building something that has the power to profoundly impact how people feel seen, heard, and understood.
We’re looking for an AI specialist who shares this passion and wants to contribute to something bigger than just another tech project. By helping us fine-tune our GPT-4 Mini model, you’ll play a critical role in creating a product that doesn’t just work—it resonates, inspires, and changes lives.
Why This Project Matters
Every interaction with our AI has the potential to help someone feel less alone, to feel deeply understood, or to navigate their day with clarity and confidence. This role goes beyond technical execution; it’s about infusing technology with humanity. By joining us, you’re helping to make high-quality, affordable coaching accessible to everyone, bridging the gap between what people need and what they can afford.
If you’re someone who is excited by the opportunity to use your skills to make a meaningful difference, this is the project for you.
What We’re Looking For
The ideal candidate has experience fine-tuning GPT models (e.g., GPT-4 Mini), a knack for clear communication, and a proven ability to deliver results efficiently. You’ll work closely with our developers to confirm data specifications and provide a structured checklist to guide the project from start to finish.
We already have a ton of gpt 4o generated responses to customer inquiries, and we want to use that data to train a mini. Its already anonymized but we will run a scan for any PII before beginning and eradicate it if found for rock solid privacy.
Availability during our working hours, 11:00 AM to 7:00 PM PST, is required for seamless communication and collaboration.
Responsibilities
You will be responsible for:
Spec Confirmation: Collaborating with our developers to confirm the exact data specification required for model fine-tuning.
Checklist and Milestone Definition: Providing a detailed checklist of milestones and requirements that clearly outline each step of the fine-tuning process, from data readiness to final model delivery.
Data Preprocessing Coordination: Ensuring the provided data matches your confirmed specifications, ready for training.
Model Fine-Tuning: Setting up and fine-tuning the GPT-4 Mini model using our data, ensuring the model meets performance benchmarks.
Optimization: Testing and iterating on the fine-tuned model to achieve desired performance metrics (e.g., accuracy, coherence, task-specific benchmarks).
Documentation and Handover: Delivering a fine-tuned model with clear, concise documentation for internal use and future reference.
Requirements
Experience: Demonstrated experience fine-tuning smaller GPT models like GPT-4 Mini or similar. Please include examples or case studies of past work.
Technical Skills:
Proficiency in Python and frameworks like Hugging Face Transformers.
Experience with model training pipelines, including tokenizer adaptation and hyperparameter tuning.
Familiarity with GPU/TPU utilization for efficient training.
Communication Skills: Ability to clearly articulate data specifications to our developers and outline requirements in a structured, milestone-driven approach.
Availability: Must be available during 11:00 AM to 7:00 PM PST for real-time collaboration.
Key Process Details
We will provide the necessary data in the format you specify, but it is your responsibility to:
Confirm Data Specifications: Work with our developers to define the exact format and structure needed for the fine-tuning process.
Submit a Milestone Checklist: Provide a detailed checklist that outlines each step of the project, including requirements for data readiness, model setup, training, and testing.
Data Validation: Ensure the provided data matches your specifications before proceeding with training.
Training and Fine-Tuning: Set up and execute the training process, monitoring progress and iterating as needed to meet benchmarks.
Handover: Submit the completed model with comprehensive documentation of the training process
Why Work With Us?
We have multiple similar projects in the pipeline, and successful completion of this project can lead to immediate opportunities for additional work. We value affordability without compromising quality and reward clear communication, structured planning, and efficient execution.
If this sounds like a great fit, we’d love to see your application! Please include:
Examples of relevant projects you’ve worked on.
Your hourly or project-based rate.
Your availability during our working hours (11:00 AM to 7:00 PM PST).
