Remote job
AI/ML Engineer
Job details
About this role
Role overview Build and improve machine learning models for complex, practical problems, taking work from initial investigation through training, evaluation, and iteration. You’ll partner with engineering and product teams to turn research ideas into usable solutions and help develop the data pipelines and tools that support experimentation and deployment.
Responsibilities - Prepare datasets and develop, train, debug, and refine machine learning models. - Design experiments, define evaluation metrics and baselines, and use results to guide technical choices. - Explore techniques such as deep learning, reinforcement learning, and large language model training or fine-tuning where they fit the problem. - Investigate model errors and performance to understand why approaches succeed or fail. - Work with engineering and product colleagues to translate research into practical solutions. - Improve data pipelines, experimentation tools, and infrastructure for scalable model development.
Requirements - Strong grounding in machine learning, deep learning, probability, statistics, algorithms, and data structures. - Hands-on experience developing and training models through research or industry work. - Familiarity with the end-to-end ML workflow, including data preparation, experimental design, evaluation, and error analysis. - Experience with a framework such as PyTorch or TensorFlow. - Strong programming skills in Python, C++, or another relevant language. - Ability to investigate technical problems independently and communicate findings clearly; a PhD in a quantitative field is preferred, and recent graduates with practical research experience are welcome.
Nice to have - Publications at leading machine learning or language technology conferences. - Experience with reinforcement learning, LLM pre-training or fine-tuning, distributed or large-scale training, or ML systems optimization. - Experience deploying models or building serving, monitoring, and MLOps systems. - Familiarity with data processing frameworks such as Spark or Flink.