Senior AI Research Engineer - NZ

As a Senior AI Research Engineer at Leonardo.Ai, you will be responsible for training and evaluating large AI models, optimizing model training processes across thousands of GPUs, and engineering infrastructure to handle complex multimodal data. Additionally, you will collaborate with various teams to enhance platform stability and user experience, while mentoring junior engineers.

Candidates for the Senior AI Research Engineer position at Leonardo.Ai should possess proficiency in Python and deep learning frameworks, with experience in managing AWS or GCP infrastructure. A background in optimizing AI model inference, machine learning algorithms, and familiarity with generative AI techniques like diffusion-based image generation and VLMs is also essential.

At Leonardo.Ai, the work environment is flexible and inclusive. Employees can choose to work remotely or from vibrant office locations in Australia. The company values work-life balance and encourages collaboration among cross-functional teams to foster creativity and innovation.

A Senior AI Research Engineer can significantly contribute to Leonardo.Ai’s mission by developing and optimizing cutting-edge AI models, mentoring junior team members, and driving the strategic planning necessary for innovative projects. Your expertise in AI will help shape the tools that democratize design and ignite creativity at a global scale.

Leonardo.Ai is committed to the professional growth of its employees. As a Senior AI Research Engineer, you will have access to continuous learning opportunities, mentorship, and the chance to work on impactful projects. The company encourages you to tailor your career growth according to your goals and aspirations.

To optimize model training at scale, I would start by analyzing the bottlenecks in the current process, such as data loading times and GPU utilization. Implementing efficient data pipelines, utilizing asynchronous data loading, and tuning hyperparameters could significantly improve training speed and efficiency.

I have extensive experience working with deep learning frameworks like TensorFlow and PyTorch. I have utilized them for various computer vision and natural language processing tasks, developing, training, and deploying models. I also have hands-on experience with optimizing model performance in production environments.

For multi-modal data processing, I would leverage architecture designs that can handle diverse data types effectively, like transformers. Additionally, creating uniform representations across modalities and employing transfer learning techniques would allow for improved performance on complex tasks.

I actively engage with AI research by reading relevant papers, following leading AI researchers on platforms like Twitter, and participating in online courses and webinars. I also contribute to open-source projects and attend industry conferences to learn about emerging trends and technologies.

In a previous project, I worked on developing a real-time image recognition system. The challenge was managing the large dataset and tuning the model parameters effectively. I implemented Agile methodologies to break tasks into manageable sprints, which helped streamline the process and allowed for regular performance assessment and adjustments.

Collaboration with cross-functional teams is essential in AI research. I make it a point to maintain open channels of communication, ensure everyone has clear roles, and set regular check-in meetings. This approach helps in aligning goals and fosters a collective problem-solving environment.

Mentoring junior engineers involves providing guidance through knowledge sharing and hands-on support. I encourage them to explore independently while being available for questions. Conducting pair programming sessions and fostering a culture of feedback are also crucial in their development.

I have extensive experience deploying AI models using cloud platforms like AWS and GCP. I focus on creating scalable APIs for model serving while implementing strategies for monitoring model performance and automating retraining processes when necessary.

Generative AI refers to algorithms that can generate new content based on learned patterns from training data. It is important for applications like creative design, content generation, and enhancing user experiences, as it fundamentally changes how we interact with technology.

I am excited about the opportunity to work with Leonardo.Ai because of its mission to democratize design and creativity through AI. I believe my skills in AI research, coupled with my passion for innovative technology, align directly with the company’s goals of driving impactful advancements in the AI space.