ML Software Engineer, LLM (2+ years experience)

Key Skills

Python Programming, Git, Machine learning techniques, Deep Learning, Keras software library, Natural Language Processing (NLP), PyTorch, TensorFlow, Docker, GPUs, JAX framework, Large Language Models - LLMs, PyG- PyTorch Geometric

Job Description

In the Engineering Team, ML Software Engineers focus on understanding business problems, identifying and harnessing the right AI or cognitive computing technologies to solve them, and take part in the commercial deployment of their solution.

ML Software Engineers intervene at every stage of the life of a project. They provide technical expertise and scope projects during pre-sales, drive research and development for Proof of Concepts (POC), develop Minimum Viable Products (MVPs), and ensure continuous improvement and maintenance for deployed solutions.

As a NLP and Large Language models (LLMs) Specialist you will manage large NLP-related projects, products, and solutions. Among the tasks a ML Software Engineer is responsible for, it is important to mention: developing prototype applications, providing software design and programming support, implementing software libraries, supporting the business development, managing projects and project teams.

In this role, at InstaDeep you will report to a Lead Research Engineer. This role may require overseas travel, ML Software Engineers must be available to travel.

ROLE RESPONSIBILITIES/ACCOUNTABILITIES

• Design, implement and deliver performant and scalable machine learning libraries.
• Write and maintain high-quality, maintainable and modular code with concise documentation and continuous integration together with engineers.
• Implement models to learn the semantics of structured and unstructured data (CVs, text, graph data, images, audio, and other modalities).
• Apply knowledge of LLMs to develop industrial solutions (POC, MVP, fully deployed solution) and products (platforms and frameworks). Apply best practices for robustness and scalability when considering LLM applications.
• Start leading small teams/projects in the year or two following recruitment.
• Put as a priority the project’s proper delivery with high quality coding practices.
• Adapt machine learning and neural network algorithms and architectures to best exploit modern parallel environments (e.g. distributed clusters, multicore SMP, and GPU).
• Report and present development work clearly and efficiently, both internally and externally (clients, different audiences, etc.), both verbally and in writing.

REQUIREMENTS:

• Postgraduate degree in Computer Science, Mathematics or a related scientific field.
• Strong interest in leading small teams/projects in the year or two following recruitment.
• Experience in building projects and products for at least 2 years.
• Experience in developing and debugging Python.
• Proficiency with Git workflow, development with best coding practices.
• Experience in deploying software to production using Docker.
• Experience with Natural Language Processing.
• Experience developing LLM based applications.
• Experience using deep learning frameworks such as PyTorch, Tensorflow, Jax and/or Keras is a plus.
• Experience leading teams or projects is a strong preference.
• Research and software engineering experience demonstrated via previous work experience, internships, contributions to open source projects, or coding competitions.