Biomedical Signal Processing Engineer

Sempulse; Town of Texas (WI); Hybrid; USD 120,000 - 170,000; Full time

Job summary

Sempulse is seeking an opportunistic Biomedical Signal Processing Engineer to design and deploy biosignal algorithms on wearable hardware. You will own signal processing for PPG/ECG data, and drive ML model development from prototyping to deployment on constrained embedded platforms.

The role emphasizes hands-on development, independent work, and collaboration with hardware, firmware, and clinical teams. Remote work with occasional travel is expected; Austin/San Antonio or nearby Texas locations

Qualifications

• BS or higher in Electrical/Biomedical/CS or related field.
• Hands-on experience developing signal processing or ML algorithms for wearable biosensor data (PPG preferred).
• Proficiency in MATLAB and Python with algorithm benchmarking.
• Embedded ML deployment on ARM Cortex-M (CMSIS-DSP/CMSIS-NN) experience.
• Strong communication, problem-solving, and self-motivation skills.

Responsibilities

• Design biosignal processing algorithms for wearable data (PPG/ECG) on ARM Cortex-M.
• Own the full ML pipeline: feature engineering, model selection, quantization.
• Develop data collection protocols with hardware/firmware teams.
• Apply rigorous statistical methods and cross-validation design.
• Prototype algorithms using MATLAB and Python; document for FH.
• Collaborate across hardware, firmware, and clinical teams for integration and verification.

Skills

Signal processing

Machine learning

Wearable biosensors

Embedded ML

Education

Bachelor's degree or higher

Graduate/Doctorate preferred

Tools

MATLAB

Python (NumPy/SciPy/Scikit-learn)

CMSIS-DSP

CMSIS-NN

Job description

Searching for an Opportunistic Biomedical Signal Processing Engineer

We are a well-funded, late-stage medical device startup with our sights set on saving lives. We have FDA-cleared, non-invasive hardware sensors that acquire patients’ vital signs anytime, anywhere. We have some great products and some great ideas, validated by the US Military, SOCOM, NATO, the Gates Foundation, the Cleveland Clinic, and many others. Our device saved its first life 2 months after our first FDA clearance and we’re now up to seven. I’d love to go into more details here, so we can find the best possible candidate, but we have confidentiality issues to contend with. That will have to wait until we sign an NDA and have a cup of coffee.

We’re looking for a hands‑on engineer to own the development of biosignal processing algorithms and their deployment onto the embedded hardware inside our wearable device. We have a clear technical roadmap, a working platform, firmware infrastructure, and a talented engineering team in place today. We also have real physiological data, validated by government-funded studies, that will feed the work. Details will have to wait until that virtual cup of coffee, but we need to convert that data and platform into production‑ready, embedded algorithms that run reliably on constrained hardware in austere environments.

You should be intelligent, creative, adaptable, and able to juggle lots of concepts at once. We’re looking for someone whose friends comment that they never forget or lose track of anything, someone who learns things quickly and retains that knowledge. A great attitude goes a long way! You must also be a problem‑solver. You must also be ambitious, self‑motivated, and equally able to work independently or in groups. You must also be an effective communicator and able to learn from and work with clients and teammates.

The ideal candidate would have knowledge and experience in wearable biosensor technology and signal processing, with exposure to the Department of War or the federal government as a plus. A bonus would be someone who comes in with thoughts and ideas about managing their tasks, working within project management concepts, and accomplishing goals and objectives.

What we’re really looking for is someone who is smart, ambitious, and can keep track of lots of things at once. You must also be able to learn quickly and apply what you’ve learned; you also need to be able to concentrate on both the forest and the individual trees at different times. Ultimately, we’re looking for aptitude and not decades of experience. We need someone to roll up their sleeves, but this is the kind of role that will most likely morph into a leadership role in the next few years. There may be some travel, but we can locate our trials to your general area over time. [TT1] [KS2]

Since we are a startup, everyone has diverse roles and no two days are ever the same. If you like to punch the clock and do the same things every day, this role is probably not for you. We have lots of great resources on staff to learn from. Our company culture is having fun while working hard to save lives.

We are well-funded with awards and budgets for this work, but the answer cannot be just licensing off‑the‑shelf algorithms. We require creativity to bridge the gap between public‑domain signal processing literature and the constraints of our specific hardware and real‑world use cases. The work must follow best practices and produce defensible, verifiable results, and we will be forward‑thinking in how we approach the engineering tradeoffs.

Confidence is good. I want someone who is confident in their abilities and doesn’t let things get in the way of the positive things they can contribute. These qualities usually lean towards a positive‑leadership style. We believe very strongly in surrounding ourselves with the best and brightest, so should you.

Responsibilities

1. 1. Design and implement biosignal processing algorithms for PPG, ECG, body temperature, and inertial sensor data, targeting deployment on ARM Cortex‑M embedded hardware.
2. 2. Own the full ML development pipeline: feature engineering, model selection and bake‑off, sensor ablation, and quantized model deployment using CMSIS‑DSP and CMSIS‑NN.
3. 3. Develop and execute data collection protocols in collaboration with the hardware and firmware teams, including multi‑subject labeling and leakage‑safe cross‑validation design.
4. 4. Apply rigorous statistical methods: whole‑participant holdout, Shapley‑value sensor ablation, pairwise significance testing, and pre‑registered decision thresholds.
5. 5. Work with MATLAB and Python (NumPy, SciPy, scikit‑learn) for algorithm prototyping, validation, and documentation in support of the design history file.
6. 6. Collaborate across hardware, firmware, and clinical teams to define sensor integration requirements, timing synchronization constraints, and verification benchmarks for production deployment.

Requirements and Qualifications

• Bachelor’s degree or higher in Electrical Engineering, Biomedical Engineering, Computer Science, Physics, Mathematics, or a closely related field. Graduate or Doctorate degree preferred.
• Hands‑on experience developing signal processing or machine learning algorithms for wearable biosensor data, particularly PPG. Prior work with photoplethysmography is strongly preferred.
• Proficiency in MATLAB and Python (NumPy, SciPy, scikit‑learn). Experience implementing and benchmarking algorithms in both environments is required.
• Demonstrated experience with embedded ML deployment on ARM Cortex‑M processors (nRF52840 or similar), including quantization, CMSIS‑DSP / CMSIS‑NN integration, and hardware‑in‑the‑loop profiling.
• Functional knowledge of statistical analyses and reporting.
• High native intelligence: fast learner, sharp pattern recognition, retains detail without being reminded.
• Genuine creativity: comfortable finding compliant, defensible ways to execute unconventional testing scenarios, not just the standard playbook.
• Working knowledge of wearable biosensor technology: PPG contact mechanics, ECG signal quality, IMU‑based motion artifact characterization, 3D spatial optimization, and multi‑sensor fusion.
• Algorithm development experience is a significant plus: rule‑based, classical ML (logistic regression, SVM, tree ensembles), and small neural models for tabular biosignal features.
• Exceptional communication and interpersonal skills: genuinely easy to work with, low‑ego, and builds trust quickly with teammates, clients, and government partners.
• Strong self‑motivation: productive and disciplined working from home without oversight.
• Excellent organization, memory, and attention to detail: nothing falls through the cracks.
• Time management and the ability to juggle multiple concurrent workstreams.
• Problem‑solving orientation, focused on business objectives over process for its own sake.
• Office 365, SharePoint, and Git proficiency.
• Professional experience using AI‑assisted research and development tools and a clear understanding of their appropriate applications and limitations.

** No offshore resources or hiring agents need respond. You would predominantly work remotely with few trips per year. The preferred location is Austin, TX or San Antonio, TX (proximity to our lab and partner facilities is a plus). Other considerations may be given to individuals located in Texas, Florida, Georgia, Idaho, Pennsylvania, Maryland, Delaware, Virginia, or California.