Software Engineer Reinforcement Learning bei DFINITY

Employment Type: 6 Month Contract

We are looking for a Software Engineer with a focus on data preparation and AI model training. You will work on assembling, annotating, and cleaning training data, while contributing to reward modeling and supervised fine-tuning tasks.

You might thrive in this role if you:

• Have a deep understanding of machine learning and machine learning applications.
• Working knowledge and experience tuning large language models (multimodal) and building evaluations.
• Be willing to dive into large codebases to debug.
• Someone who thrives in a dynamic and technically complex environment.
• Track record of delivering outside-the-box novel solutions to solve real-world constraints.

Responsibilities

• Data Assembly & Annotation: Gather and annotate training data for AI models, ensuring it meets the quality requirements for reward modeling and supervised fine-tuning.
• Data Cleaning & Processing: Conduct data cleaning and preprocessing to ensure models receive high-quality input.
• Model Training: Participate in the training and fine-tuning of models, ensuring that they meet performance and accuracy standards.
• Collaboration: Work with AI engineers, data scientists, and other team members to ensure efficient workflows and data handling.
• Continuous Improvement: Support iterative improvements to models based on performance monitoring and feedback.

Requirements

• Experience: At least 3 years of experience working in a software engineering role focused on AI/ML tasks.
• Data Expertise: Hands-on experience assembling, annotating, and cleaning training data for machine learning models.
• Technical Skills: Proficiency in Python and experience with AI frameworks like TensorFlow or PyTorch.
• Model Training: Familiarity with model training, reward modeling, and supervised fine-tuning techniques.
• Attention to Detail: Strong focus on data quality and attention to detail when handling large datasets.

Bonus Points

• Experience working with reward modeling for AI systems.
• Familiarity with data labeling tools and techniques for supervised fine-tuning.
• Knowledge of cloud platforms for AI/ML workloads.