AI Working Student (gn)

at  castLabs GmbH

Remote or Berlin Office (Flexible)
Position Type: Part-time (20 hours per week) flexible work schedule
Step into an exciting opportunity as our new AI Student, welcoming applicants of all genders! This role is your gateway to cutting-edge AI technologies, where you’ll help build and refine AI-powered solutions to drive innovation across multiple product lines.

ABOUT US:

At castLabs, we pioneer innovative solutions in video streaming software, cloud services, and content security. Our forensic watermarking product leverages advanced technologies to safeguard media content integrity and traceability. We are looking for an AI student to support the development and training of specialized computer vision models tailored to enhance the capabilities of our forensic watermarking system. This is a hands-on product development role, not a research-focused position.

PROFILE REQUIREMENTS:

• Currently pursuing a Master’s degree in Artificial Intelligence, Machine Learning, Computer Science, or a related field.
• Proficiency in Python and experience with frameworks such as PyTorch or TensorFlow.
• Solid understanding of machine learning principles and foundational con cepts:
• Convolutions, transformers, fully connected layers, residual connections, etc.
• Backpropagation, loss functions, regression/classification tasks, supervised/unsupervised learning methods.
• Problem-solving mindset with the ability to work independently and deliver results.
• Strong communication and teamwork skills.

PREFERRED QUALIFICATIONS:

• Hands-on experience in AI model deployment and MLOps practices.
• Background in computer vision technologies and model optimization.
• Familiarity with tools like PyTorch Lightning or Keras.
• Knowledge of database systems and APIs for data storage/retrieval.
• Proficiency with Git and collaborative coding platforms like GitHub or GitLab.
• Interest or experience in media technology solutions or AI-driven tools.

KEY RESPONSIBILITIES:

• Develop and train computer vision models tailored to forensic watermarking applications.
• Collaborate with the Engineering and Product teams to define model objectives, training data requirements, and evaluation criteria.
• Generate, preprocess, and validate high-quality training data.
• Design and implement machine learning workflows for model training, evaluation, and optimization.
• Analyze model performance and propose actionable enhancements.
• Integrate the models into existing workflows and assess scalability.
• Document development processes, methodologies, and key findings.