What you get to do in this role:
- Generate and evaluate synthetic data tailored to improve the robustness, performance, and safety of machine learning models, particularly large language models (LLMs).
- Train and fine-tune models using curated datasets, optimizing for performance, reliability, and scalability.
- Design and implement evaluation metrics to rigorously measure and monitor model quality, safety, and effectiveness.
- Conduct experiments to validate model behavior and improve generalization across diverse use cases.
- Collaborate with engineering and research teams to identify risks and recommend AI safety mitigation strategies.
- Participate in the development, deployment, and continuous improvement of end-to-end AI solutions.
- Contribute to architectural and technology decisions related to AI infrastructure, frameworks, and tooling.
- Promote modern engineering practices including continuous integration, continuous delivery, and containerized workflows.
Qualifications
Key qualifications:
- 5+ years of experience in machine learning, deep learning, and AI systems.
- Proficiency in Python and frameworks like PyTorch, TensorFlow, and NumPy.
- Experience in synthetic data generation, model training, and evaluation in real-world environments.
- Solid understanding of LLM fine-tuning, prompting, and robustness techniques.
- Knowledge of AI safety principles and experience identifying and mitigating model risks.
- Hands-on experience deploying and optimizing models using platforms such as Triton Inference Server.
- Familiarity with CI/CD, automated testing, and container orchestration tools like Docker and Kubernetes.