Axon AI - Senior Machine Learning Scientist

As a Senior Machine Learning Scientist at Axon, your primary responsibilities include researching and developing advanced Multimodal Large Language Models (MLLMs) and computer vision techniques applicable to our cloud and devices. You'll design scalable models for real-world applications, optimize algorithms for resource-constrained environments, and collaborate with cross-functional teams for deployment and integration.

To qualify for the Senior Machine Learning Scientist role at Axon, you need a PhD with over 5 years of experience in computer science or a related field. A proven track record in research, especially in machine learning and computer vision published in reputable journals, is essential. Proficiency in programming languages such as Python and experience with deep learning frameworks like TensorFlow or PyTorch are also required.

Axon supports the growth of its Senior Machine Learning Scientists through diverse learning and development programs designed to enhance your skills and knowledge. You'll also have opportunities to contribute to academic publications and engage in technical documentation, ensuring you stay at the forefront of the field while sharing insights with the broader community.

At Axon, a Senior Machine Learning Scientist significantly impacts our mission to Protect Life by advancing technologies that improve safety and justice. Your work on MLLMs and computer vision will facilitate intelligent data perception, fostering innovations that directly contribute to societal enhancement.

The work environment for a Senior Machine Learning Scientist at Axon is collaborative and fast-paced. You'll be part of a supportive team passionate about driving meaningful change. We value diverse perspectives and provide a space where innovation thrives, encouraging open communication and teamwork on exciting projects.

When discussing your experience with computer vision algorithms, focus on specific projects where you applied techniques like object detection or scene understanding. Mention the challenges you faced, how you optimized algorithms for performance, and the results achieved, showcasing your technical expertise.

In developing MLLMs, I assess the datasets available, considering their quality and relevance. I prioritize establishing a suitable architecture, experiment with modeling approaches, and iterate based on evaluation metrics. It's crucial to communicate findings clearly to ensure alignment with team objectives.

To stay current in AI and machine learning, I regularly browse top-tier journals, attend relevant conferences, and participate in online communities. I also engage with open-source projects, ensuring I am applying and testing the latest research in practice.

In a previous project, I optimized a deep learning model for deployment on mobile devices by simplifying the architecture and quantizing the model. This reduced the memory footprint and enhanced inference time, enabling efficient performance on constrained hardware.

Mentoring junior scientists involves providing structured guidance and support. I focus on encouraging critical thinking, availability for questions, and offering constructive feedback on their projects, ensuring they feel confident in contributing to the research while also learning continuously.

I am proficient in Python and C++, which I have used extensively in data manipulation, algorithm implementation, and building machine learning models. I leverage libraries like TensorFlow and PyTorch to streamline development cycles and enhance model performance.

In my last role, I co-authored a paper on novel approaches to object recognition in surveillance environments. I contributed through extensive research, drafting sections of the manuscript, and presenting our findings at a major conference, which fostered discussions about our work with peers.

Integrating models into production systems often presents challenges such as ensuring compatibility with existing infrastructure and dealing with unexpected data patterns. I tackle these by conducting thorough testing and iteration cycles and collaborating closely with engineering teams to address integration issues promptly.

I evaluate model performance using various metrics like accuracy, precision, recall, and F1-score for classification tasks, alongside ROC-AUC for binary classifications. I also conduct cross-validation to validate the robustness of models across different datasets.

My motivation stems from a passion for technology’s potential to solve real-world problems. Working in machine learning and AI allows me to contribute to impactful projects that advance human capabilities, such as safety and security, which aligns perfectly with Axon’s mission.