Model Implementation Engineer

Sciforium is an AI infrastructure company developing next-generation multimodal AI models and a proprietary, high-efficiency serving platform. Backed by multi-million-dollar funding and direct sponsorship from AMD with hands-on support from AMD engineers the team is scaling rapidly to build the full stack powering frontier AI models and real-time applications.

About the role

We are seeking a highly skilled Model Implementation Engineer who is passionate about bringing cutting-edge machine learning models into production-ready systems. In this role, you will implement, maintain, and optimize a large and evolving library of state-of-the-art models across modalities, ensuring high performance and reliability from day one.

You will work at the intersection of research and systems, translating the latest ideas into robust, scalable implementations. This includes collaborating closely with GPU kernel and systems teams to ensure models are efficiently executed on modern accelerators.

This role is ideal for someone who thrives in fast-moving environments, enjoys working across a wide range of model architectures, and wants to play a key role in enabling rapid adoption of the latest advancements in AI.

Key Responsibilities

• Maintain and evolve a large-scale library of modern machine learning models, including but not limited to LLMs, ASR, TTS, image and video models, and diffusion-based systems.

• Implement new model architectures and research ideas, ensuring correctness, scalability, and production readiness.

• Rapidly integrate newly released open-source models to enable day-0 support across the platform.

• Collaborate closely with GPU kernel and systems teams to optimize model execution and improve overall performance.

• Benchmark models rigorously and ensure they meet internal performance, latency, and efficiency standards.

• Contribute to the canonicalization and standardization of model implementations across the library.

• Develop and maintain internal tooling, testing frameworks, and documentation to support model reliability and reproducibility.

Must-Haves

• At least 3 years of industry or research experience in model implementation or applied machine learning.

• Master of Science (or higher) in Computer Science, Machine Learning, Electrical Engineering, Applied Mathematics, or a related field.

• Strong programming skills in Python and experience working with modern ML frameworks.

• Hands-on experience with JAX and/or PyTorch (JAX strongly preferred).

• Proven experience maintaining and developing model libraries or reusable ML components.

• Solid understanding of deep learning architectures across multiple domains (e.g., NLP, vision, speech, generative models).

• Experience implementing models from research papers and adapting them for real-world usage.

• Ability to work across teams and collaborate with systems and performance engineering groups

Nice-to-Haves

• Experience with model performance optimization and profiling.

• Familiarity with low-level performance considerations when running models on GPUs/TPUs.

• Experience working with large-scale model training or inference systems.

• Contributions to open-source model repositories or ML frameworks.

• Experience with JAX-first workflows and advanced features (e.g., pjit, xmap, or custom transformations).

Benefits include

• Medical, dental, and vision insurance

• 401k plan

• Daily lunch, snacks, and beverages

• Flexible time off

• Competitive salary and equity

Equal opportunity

Sciforium is an equal opportunity employer. All applicants will be considered for employment without attention to race, color, religion, sex, sexual orientation, gender identity, national origin, veteran or disability status.