Job Url: https://www.linkedin.com/jobs/search/?currentJobId=4368112836&distance=25.0&f_AL=true&f_TPR=r86400&f_WT=2&geoId=103644278&keywords=software%20engineer&origin=JOB_SEARCH_PAGE_JOB_FILTER&spellCorrectionEnabled=true&start=25 Job Description: Senior Software Engineer Wavelet Medical · United States (Remote) Easy Apply Save Save Senior Software Engineer at Wavelet Medical Show more options Your profile is missing required qualifications Show match details Help me update my profile BETA Is this information helpful? Get personalized tips to stand out to hirers Find jobs where you’re a top applicant and tailor your resume with the help of AI. Try Premium for PKR0 About the job About Wavelet Wavelet is building the first non-invasive electroencephalography (EEG) system to measure fetal brain activity during pregnancy and labor. Our technology combines novel sensing hardware with advanced signal processing and machine learning to extract clinically meaningful EEG signals from complex biological noise. Our mission is to reduce preventable brain injury at birth. We are a venture-backed, early-stage medtech company advancing toward FDA clearance of a Class II medical device. Senior / Staff Algorithm Engineer (Signal Processing + ML + Physiologically Informed Modeling) Wavelet is seeking a senior, hands-on engineer to work on core algorithm development for our fetal EEG system. This role focuses on designing and validating robust methods for low-SNR signal reconstruction and state/event detection in complex biological recordings, and building the supporting modeling and evaluation infrastructure needed to make those methods reliable across real-world variability. You do not need prior expertise in fetal medicine. We’re looking for a strong technical builder who can apply fundamentals in signal processing, machine learning, and modeling/simulation to a high-impact biomedical problem. What You’ll Do Develop and improve algorithms for extracting targeted brain signals from multichannel abdominal recordings, including denoising, separation/reconstruction, and robustness to nonstationary artifacts. Develop and validate event/state detection methods relevant to clinically meaningful neurophysiology patterns (e.g., suppression-like regimes), with explicit attention to false alarms, timing, and uncertainty/quality gating. Build and maintain physiologically informed modeling tools that support algorithm development and evaluation, including: forward/physics-based modeling workflows (e.g., FEM toolchains and parameter sweeps), incorporation of real physiological time series priors/templates (e.g., neonatal EEG), realistic artifact and interference models (motion, impedance variability, cardiac contamination, baseline drift, sensor noise), and controlled perturbations/morphology transforms to probe algorithm failure modes and robustness. Establish rigorous evaluation practices: leakage control, locked splits, stratified performance reporting, and reproducible experiment workflows (versioned data, configs, model artifacts). Collaborate closely with hardware and software engineers to align algorithm assumptions with acquisition realities (SNR, channel integrity, dropout, impedance changes) and to integrate algorithms into repeatable offline workflows, with a path toward eventual real-time constraints. Contribute to technical documentation and test/validation habits appropriate for a regulated medtech environment (clear artifacts, traceability, verification-friendly code). What Success Looks Like (first 6–12 months) A reproducible end-to-end analysis workflow that runs reliably on real sessions and produces consistent outputs with clearly documented limitations and failure modes. A modeling/evaluation framework that can systematically stress-test algorithms under controlled physiological and artifact variability, and quantify improvements. Clear experimental discipline across the team: repeatable runs, versioned artifacts, and defensible metrics. Required Qualifications MS/PhD in EE, BME, applied math, CS, physics, or similar, or experience developing software in a highly regulated and quality-driven environment. Strong Python engineering skills; ability to build maintainable, testable algorithm codebases. Experience developing algorithms for noisy time series (biomedical signals preferred but not required) and handling nonstationarity/artifacts. Practical ML experience (PyTorch preferred): training/debugging, data pipelines, reproducibility. Strong signal processing fundamentals (filtering, time-frequency, multichannel methods, robust estimation). Experience with modeling/simulation workflows sufficient to run and sanity-check forward models and automate sweeps (FEM toolchains such as Sim4Life/ZMT, COMSOL, ANSYS, or comparable). Strong experimental discipline: leakage prevention, proper splits, subgroup analyses, and correctly interpreted metrics. Comfortable operating in an early-stage startup environment with evolving requirements and high ownership. Preferred Qualifications (nice-to-have) Inverse problems / source separation / blind source separation. Event detection / segmentation / sequence modeling for noisy time series. Experience building artifact models or realistic perturbation frameworks for physiological signals. Uncertainty quantification/calibration; quality gating strategies. GPU/HPC experience for large sweeps (profiling, distributed training, cluster automation). Familiarity with regulated development habits (design controls, verification evidence organization; IEC 62304 awareness is a plus). What We Offer Competitive salary and meaningful equity High ownership and technical autonomy Close collaboration with clinicians, engineers, and researchers Flexible work arrangements