Perceptra Technologies has raised €1.2 million (approximately $1.3 million) from photonics accelerator PhotonDelta [PhotonDelta, 2026]. The Oakland-based startup is betting that a chip-scale optical sensor, paired with machine learning, can give biomanufacturers a real-time read on what’s happening inside their bioreactors. The goal is to replace the slow, offline assays that currently dominate biologics production with a continuous data stream.
Founded in 2021, the company is led by a team of MIT researchers and hardware veterans. Their pre-seed funding, which also includes a $275,000 SBIR award from the National Science Foundation [SBIR.gov], is earmarked for developing a benchtop analyzer and photonic biosensor chips. The core bet is that shrinking a Raman spectrometer onto a silicon photonic integrated circuit (PIC) will unlock the speed and multiplexing needed for inline process monitoring.
The hardware wedge
Perceptra’s proposed wedge is a hardware-software bundle. The hardware is a compact Raman analyzer built on proprietary photonic chips. The software is a suite of AI models trained to interpret spectral data and predict concentrations of key biomolecules like glucose and ethanol [Product | Perceptra]. The company claims its first-generation benchtop analyzer can achieve a 1 g/L detection limit and requires only five data points for initial calibration, a significant reduction from traditional methods.
The value proposition targets a specific pain point: the lag between sampling and results in biopharma. Current practice often involves pulling a sample from a bioreactor and sending it to a lab for analysis, a process that can take hours. Perceptra positions its platform as an inline or at-line alternative, promising higher multiplexing, faster response, and lower sample volumes than incumbent spectroscopic or wet lab assays [Perplexity Sonar Pro Brief].
A team built for deep tech
The founding team’s credentials are squarely in the deep-tech lane. The technical co-founders hail from MIT’s integrated photonics and hardware design circles, while the advisory board includes industry operators.
| Role | Name | Prior Affiliation | Notable Credential |
|---|---|---|---|
| Co-Founder & CEO | Amir Atabaki | MIT Research Scientist | 12+ years in chip-scale biosensors & spectroscopy [Perplexity Sonar Pro Brief] |
| Co-Founder & CTO | Daniel Kramnik | Google X (Project Taara), Astranis | Led ASIC design for optical wireless comms; PhD from UC Berkeley [Daniel Kramnik, PhD - Perceptra Technologies, 2026] |
| Co-Founder & Scientific Advisor | Rajeev Ram | MIT Professor, Partner at Breakthrough Energy Ventures | Co-founded Ayar Labs and Erbi Biosystems (acquired by Merck) [Perplexity Sonar Pro Brief] |
| Advisor | Hamid Mehdizadeh | Bristol Myers Squibb | Director of Manufacturing Intelligence [Perplexity Sonar Pro Brief] |
Rajeev Ram’s track record is particularly relevant. His prior company, Erbi Biosystems, was focused on bioprocess instrumentation and was acquired by Merck, mapping directly onto Perceptra’s target market. Hamid Mehdizadeh’s role at Bristol Myers Squibb, where he leads integration of sensor and AI solutions in manufacturing, provides a potential conduit for industry feedback.
Why PhotonDelta wrote the check
The €1.2 million investment from PhotonDelta is a strategic, sector-specific vote of confidence. PhotonDelta is a European accelerator focused on building the integrated photonics ecosystem. Its investment followed Perceptra’s victory in the 2025 Global Photonics Engineering Contest, where the startup’s ultra-compact photonic chip-based Raman sensor won top honors [22, 24].
For PhotonDelta, Perceptra represents an application of photonics in a high-value, regulated industry. The funding is likely non-dilutive or convertible, fitting the accelerator’s model of supporting early-stage hardware development. It provides Perceptra with capital that understands the long development cycles and technical risk inherent in building novel scientific instruments.
The path to the cleanroom
The immediate next steps are clear: refine the prototype and secure pilot deployments. The company lists its primary applications as upstream bioreactor monitoring, downstream chromatography, and Process Analytical Technology (PAT) for GMP manufacturing [Applications | Perceptra]. The initial customer target is process development and manufacturing science teams at biopharmaceutical companies and contract development and manufacturing organizations (CDMOs).
Perceptra’s published roadmap suggests a focus on the “Bioprocess 4.0” trend toward smart, data-driven factories. The platform’s promised real-time data is positioned as a key enabler for automated process control, which could improve yields and reduce batch failures. The technical risk is high, but the team’s academic and industrial roots in both photonics and bioprocess instrumentation are a mitigating factor.
The integration hurdle
Even with a working sensor, commercial adoption faces significant hurdles. Biopharmaceutical manufacturing is a conservative, validation-heavy environment. Introducing a new, inline analytical instrument requires rigorous testing and regulatory alignment. Perceptra’s platform must demonstrate not just accuracy, but also robustness, reproducibility, and smooth integration into existing quality control and data management systems.
The competitive landscape, while not naming direct peers, is occupied by established providers of spectroscopic equipment and analytical services. Perceptra’s answer rests on its proprietary photonic chip, which aims to drastically reduce the size, cost, and complexity of Raman analysis. The company’s early traction signals,the NSF grant, the photonics award, and the PhotonDelta investment,are all technical validation milestones, not commercial ones. The next twelve months will be about translating that technical credibility into a paid pilot with a named biopharma partner.
Perceptra’s €1.2 million pre-seed round from PhotonDelta, layered on its NSF SBIR grant, gives it a runway to tackle that challenge. The bet is that a chip-scale photonic sensor, built by a team with pedigree in both startups and academia, can find a slot on the bench and eventually inside the bioreactor skid. The question for 2025 is which manufacturer will be first to test it in a live process.
Sources
- [PhotonDelta, 2026] PhotonDelta invests in Perceptra Technologies | https://www.perceptra.io/
- [SBIR.gov] SBIR Award Listing for Perceptra Technologies | https://www.sbir.gov/
- [Product | Perceptra] Perceptra Technologies Product Specifications | https://www.perceptra.io/product
- [Perplexity Sonar Pro Brief] Company Overview and Team Background | https://www.perceptra.io/team
- [22, 24] 2025 Global Photonics Engineering Contest Winner Announcement | https://www.perceptra.io/
- [Daniel Kramnik, PhD - Perceptra Technologies, 2026] Team Profile Page | https://www.perceptra.io/team
- [Applications | Perceptra] Target Applications for Biosensing Platform | https://www.perceptra.io/programs