Biosensor Architecture That Performs in the Real World
Expert biosensor design consulting and independent verification services. From validating the performance of a novel sensor or one of its components, to selecting the right transduction mechanism for your target, to optimizing signal gain, to engineering out noise, and scaling for manufacturability, we give you the technical clarity needed to deploy with confidence.
Where Expertise Meets Engineering
Architecture-to-Target Matching
The right sensing architecture depends on the target analyte, the sample matrix, the testing frequency, the performance requirements, and more. The wrong choices early on compound every downstream problem. I help teams evaluate and select the architecture most likely to succeed before significant resources are committed.
Design for Manufacturability
A biosensor that works in the lab but can't be made consistently at scale isn't a product. I work with development teams to identify manufacturability constraints early and align design choices with process capabilities to reduce the gap between prototype performance and product performance.
Limit-of-Detection Optimization
Reaching the required LOD is often the central challenge in biosensor development. I bring a systematic, mechanistic approach to signal amplification, background suppression, and assay optimization. My approach is grounded in the physics of the system and the measurement, not just empirical iteration.
Noise & Interference Mitigation
Real-world samples are complex. Non-specific binding, matrix effects, electrochemical interferents, and environmental variables can all degrade performance in ways that aren't visible in clean buffer or model systems. This is often one of the hardest stages of development to address because the root causes are frequently unknown, and the challenges typically arrive late in development. I help teams anticipate, characterize, and engineer around these challenges.
Independent verification for any sensing platform
Assays that work perfectly on the bench often fail during scale-up or tech transfer. This failure risks eroding trust, delaying timelines, or both. Third-party verification eliminates this risk by providing objective proof of your system's robustness. Whether you are advancing internal development or preparing for an external tech transfer, I engineer custom validation protocols driven by regulatory requirements, tailored to your exact analyte and matrix. This ensures your performance claims hold up under scrutiny, preserving stakeholder confidence and keeping your commercialization timeline on track.
What every verification covers
Intended use and acceptance criteria
Define what the sensor must do, in which matrix, and what counts as success. We will define the test procedures and pass/fail requirements in a written protocol before any testing begins.
Detection capability
Quantify a limit of detection, limit of quantification, and/or a decision cutoff point, depending on whether your sensor reports a number, a category, or a yes/no result.
Selectivity and interference
Testing against structurally related compounds, matrix components, and the sample variables that matter for your application, including multiple individual sample sources.
Precision and reproducibility
Repeatability and between-run reproducibility across the claimed range, including within-sensor, sensor-to-sensor and lot-to-lot variation.
Stability and robustness
Shelf life, in-use stability, and sensitivity to temperature, timing, and handling, using conditions that match real-world use.
Reference method comparison
Agreement with an established method such as chromatography, a validated immunoassay, or standard addition using a tracable standard, so your results are anchored to a reliable ground truth.
Tailored to your sensing modality
The framework stays the same, but what gets tested and how results are expressed changes with the technology. Three examples:
Electrochemical aptamer-based sensing
- Calibration and signal gain in the intended matrix and at the intended temperature
- Dynamic range, aptamer affinity, response time, and reversibility
- Baseline drift and fouling over the intended measurement duration, including how well drift correction works
- Sensor-to-sensor reproducibility across fabrication batches
Qualitative and semi-quantitative colorimetric sensing
- Behavior of samples near the cutoff, to map the uncertain zone between clear negatives and clear positives
- False-positive and false-negative rates against a reference method
- Agreement of color-scale categories with reference concentrations, where appropriate
- Readout variability appropiate to your platform, including the effect of variation in lighting, smartphone cameras, and color stability over the read window
Fluorescence sensing
- Detection and quantification limits based on blank variability in the real matrix
- Matrix autofluorescence, quenching, and inner-filter effects
- Photobleaching and signal stability during readout
- Instrument-to-instrument and reader-setting sensitivity, plus linear range and saturation
What you receive
- A mutually agreed upon written validation protocol before any testing begins
- A verification report that presents all results, including any that fall outside the predefined criteria
- Records of reference standards, reagent and sensor lots, and sample handling sufficient detailed to reconstruct the study
Study designs draw on established validation principles, such as FDA's ICH M10 and biomarker bioanalytical method validation guidances and CLSI evaluation protocols, selected to fit your target, modality, and intended use. These guidances are nonbinding and were not written for every sensor type, so I adapt them rather than apply them literally. Verification results describe how your sensor performed. They are not an FDA clearance, approval, or endorsement.
Discuss VerificationRigorous by design.
Practical by necessity.
I’m Dr. Herschel Watkins. I earned a Ph.D. in Bioengineering from the University of California, Santa Barbara in 2013. I’ve spent my career turning ideas into medical devices, with a particular focus on continuous and point-of-care biosensors. Each of the devices I've launched is the culmination of many hard-earned lessons, design changes, backtracks, and eureka moments. Every time through the cycle I’ve been able to apply lessons learned from previous launches.
Biosensor development sits at the intersection of surface chemistry, electrochemistry, biophysics, materials science, and manufacturing. Problems rarely respect disciplinary boundaries. I started Fidelity Sensors so you can leverage my experience across all these disciplines to get your sensor to market faster.
I work with teams at all stages, from early feasibility and architecture selection through to troubleshooting performance gaps in late-stage development. The goal is always the same: a sensor that works reliably, can be made consistently, and meets the performance requirements that actually matter.
Engagements are structured around your specific challenge, whether that's a focused technical review, independent verification services, ongoing advisory support, fractional technical leadership, or hands-on collaboration with your development team.
Ready to talk about your biosensor challenge?
Whether you're evaluating architectures, troubleshooting a performance issue, planning a development program, or just want help validating an existing sensor, a conversation is a good place to start. The first consultation is free.
Book a 30-minute call to discuss your project. No commitment required.
Book a Consultation