About Electrokare

Electrokare builds software that turns ECG signals into continuous, non-invasive physiological insight.

We develop machine-learning models that estimate blood-level physiology from cardiovascular signals, enabling real-time insight without lab tests or invasive sensors. Our systems are already in production, backed by proprietary datasets, and used across high-performance and medical-adjacent settings.

This is not a toy ML problem. The signals are noisy, the biology is complex, and correctness matters.

The Role

This is a core applied ML role focused on building and improving models that map ECG signals to blood-level physiological values.

You will sit at the intersection of signal processing, data science, and deep learning, working directly with the CTO to develop models that are already live and evolving toward even better performance.

The role is roughly 50% applied research and 50% production ML. You’ll read and dissect papers, design experiments, train and evaluate models, and then deploy them into real systems.

What You'll Do

• Design, train, and improve ML and deep-learning models for ECG-to-blood inference
• Work with large proprietary physiological datasets and real-world signal noise
• Apply signal processing techniques alongside modern deep-learning approaches
• Read, evaluate, and implement ideas from current scientific literature
• Move models from research to production (training → validation → deployment)
• Collaborate closely with the CTO on model architecture, evaluation, and roadmap
• Build tooling for experimentation, validation, and performance tracking
• Communicate results clearly through metrics, visualizations, and reports

Technical Requirements:

Machine Learning & Data

• Deep learning with TensorFlow/Keras and/or PyTorch
• Strong foundation in signal processing and time-series analysis
• Solid data science fundamentals and statistical reasoning

Programming

• Proficiency in Python
• Experience with one or more of: C++, Rust, R, C
• Ability to read and work across multiple languages as needed

Infrastructure & Deployment

• Model deployment using Docker and Kubernetes
• Cloud experience (GCP and/or Azure)
• Working knowledge of databases (SQL, MongoDB, Bigtable, etc.)

Communication

• Clear data visualization and reporting
• Ability to explain complex models and results to technical and non-technical teammates

AI Tools

• Comfortable using modern AI-assisted development tools (e.g., Claude Code, CodeX, similar)
• Uses tools to move faster, not to replace judgment

Research

• Strong ability to read, understand, and critically evaluate scientific papers
• Experience implementing methods from literature, not just using libraries
• Background in applied research, with publications strongly preferred

Ideal Background

• MS or PhD in machine learning, computer science, electrical engineering, applied math, biomedical engineering, or related field
• 3-7 years of experience
• Prior work on physiological signals, biosignals, medical ML, or similar domains is a strong plus
• Experience working in environments where models are deployed and used, not just published

Culture Fit

• Low ego, high ownership
• You like being useful
• You care about doing things well, even when no one’s watching
• You’re comfortable jumping in wherever needed
• You care about correctness, not just accuracy metrics
• You’re comfortable working with messy real-world data
• You’re curious and skeptical in the right ways
• You want your models to matter in practice
• You work VERY hard