AI/ML Engineer

Tiposi; Milpitas (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

Tiposi in Milpitas, California, is building applied machine learning for medical imaging from noisy, physics-constrained sensor data. You will develop reliable models and production pipelines for image reconstruction and stroke detection, collaborating with hardware, signal processing, and clinical teams.

The role emphasizes robust system-level design, handling real-world data, and integrating with DSP and software engineers to deploy validated solutions.

Qualifications

• M.S. in Computer Science, Electrical Engineering, or related field, or 5+ years of industry ML experience.
• Experience applying diffusion models, VAEs, or encoder–decoder architectures to 2D or 3D data.
• Experience building and maintaining ML pipelines beyond notebooks, including work with noisy, imperfect real-world data.
• Solid foundations in linear algebra, probability, and optimization, with the ability to reason about system-level constraints.
• Experience in reconstruction, inverse problems, physics-informed ML, or robustness and uncertainty estimation is relevant.
• Signal/image processing, multimodal modeling, GPU/edge deployment, or medical device and regulated healthcare experience is preferred.

Responsibilities

• Train multi-task models for stroke detection using RF-derived features, including classification and auxiliary prediction tasks.
• Build and maintain production-quality ML pipelines for signal-to-image reconstruction.
• Evaluate approaches to inverse problems under noise, limited data, and physical system constraints.
• Integrate models with signal processing and hardware pipelines alongside DSP, hardware, and software engineers.
• Debug training failures, data problems, and edge cases to improve model reliability and robustness.
• Balance model complexity, interpretability, and performance, making choices that support dependable system-level behavior.

Skills

Diffusion models

VAEs

Encoder–decoder architectures

Noisy real-world data handling

Education

M.S. in Computer Science or Electrical Engineering

Job description

Build applied machine learning for medical imaging from noisy, physics-constrained sensor data. Develop reliable models and production pipelines for image reconstruction and stroke detection, working within real hardware, signal processing, and clinical constraints.

What you will do

• Train multi-task models for stroke detection using RF-derived features, including classification and auxiliary prediction tasks.
• Build and maintain production-quality ML pipelines for signal-to-image reconstruction.
• Evaluate approaches to inverse problems under noise, limited data, and physical system constraints.
• Integrate models with signal processing and hardware pipelines alongside DSP, hardware, and software engineers.
• Debug training failures, data problems, and edge cases to improve model reliability and robustness.
• Balance model complexity, interpretability, and performance, making choices that support dependable system-level behavior.

What we're looking for

• M.S. in Computer Science, Electrical Engineering, or a related field, or 5+ years of industry experience in machine learning.
• Experience applying diffusion models, VAEs, or encoder-decoder architectures to 2D or 3D data.
• Experience building and maintaining ML pipelines beyond notebooks, including work with noisy, imperfect real-world data.
• Solid foundations in linear algebra, probability, and optimization, with the ability to reason about system-level constraints.
• Experience in reconstruction, inverse problems, physics-informed ML, or robustness and uncertainty estimation is relevant.
• Signal/image processing, multimodal modeling, GPU/edge deployment, or medical device and regulated healthcare experience is preferred.