☰

The EVERY Company

Scaling precision-fermented egg protein to retail volume

Industry
Precision Fermentation (Animal-Free Proteins)
Headquarters
Daly City, California, United States
Public information as of
March 2026

A4BEE prepared this analysis from publicly available sources. It reflects our own reading of The EVERY Company's published strategy and is not endorsed by, or produced in cooperation with, The EVERY Company. Company website

Strategic priorities

The EVERY Company makes animal-free egg-white protein (recombinant ovalbumin, the OvoPro™ product line) and related proteins using precision fermentation of yeast (*Komagataella phaffii*). The company began life as Clara Foods in 2014 and rebranded in 2023. In November 2025 it closed a $55 million Series D led by McWin Capital Partners to expand manufacturing capacity for a nationwide retail rollout that began at Walmart in the same month, taking cumulative funding across all rounds to roughly $290 million.

Commercial production currently runs through a network of third-party partners in Europe and Asia, including an early-stage alliance with AB InBev's BioBrew precision-fermentation division that adapts brewing infrastructure to alternative-protein output. R&D, ML model development and the core benchtop fermentation programme sit at the company's 689 Bryant St. laboratory in San Francisco. The CTO, Bart Haverkorn van Rijsewijk, joined from Zymergen and has publicly cited Zymergen's collapse as a cautionary tale about scaling biomanufacturing without an integrated data infrastructure.

Two greenfield facilities are in feasibility. A $2 million grant awarded by the U.S. Department of Defense in December 2024 funds a techno-economic feasibility assessment for a domestic biomanufacturing site, with up to $100 million of subsequent infrastructure funding contingent on the assessment. A partnership with the Abu Dhabi Investment Office (ADIO) and Vivici announced in late 2025 explores a multi-tenant 4-million-liter precision-fermentation facility in the UAE.

FDA Generally Recognized As Safe (GRAS) status is held for the current protein portfolio. The company is defending US Patent No. 12,096,784 in litigation brought by competitor Onego Bio, and supplies commercial volumes to enterprise partners including Unilever (The Vegetarian Butcher line), Grupo Palacios and Walmart.

Challenges we see

  • Digital Integration

    Connecting Daly City R&D to fermenters running in partner sites

    R&D and ML model development sit at the company's 689 Bryant St. laboratory in San Francisco. Commercial production runs at third-party partner sites in Europe and Asia, including the AB InBev BioBrew precision-fermentation network, where legacy supervisory control and data acquisition (SCADA) systems and programmable logic controllers (PLC) were not designed for secure cloud connectivity.

    Where the people who train the ML model cannot see the fermenter telemetry, the data the model is trained on is older and less rich than the data the model needs; instrumenting the line and bringing the signals into a shared data lake closes the gap before the next commercial batch.

  • Operations Manufacturing

    Translating 5-litre benchtop runs to 100,000-litre commercial fermenters

    Benchtop fermentation runs at 5 litres in San Francisco have to be translated to commercial fermenters of around 100,000 litres at partner sites. At industrial scale, mechanical shear stress, heat removal and oxygen mass transfer each behave differently than they do in a laboratory vessel, which is the route through which protein expression is suppressed at the larger volume.

    When the physical regime of the vessel changes, the assumptions built into a 5-litre recipe stop transferring by default; running the same strain at benchtop and at pilot under monitored conditions produces the data that lets the scale-up be guided rather than assumed.

  • Operations Manufacturing

    Designing US and Abu Dhabi facilities before ground is broken

    The DoD feasibility assessment for the U.S. domestic site runs through 2026, and the ADIO/Vivici Abu Dhabi facility is in early-stage planning. Both depend on precise capacity forecasting, piping topography, utility load balancing and equipment procurement choices before construction is committed.

    Where the design of a greenfield facility is frozen at the procurement stage, every later retrofit pays for the decisions taken at that point; simulating the proposed bioreactors and downstream processing lines against fluid dynamics and thermal loads surfaces the constraints before they become concrete.

  • Digital Operations

    Reading terabytes of fermenter telemetry as one dataset

    Modern bioreactors equipped with advanced sensors generate terabytes of continuous time-series data on dissolved oxygen, pH, temperature, off-gas composition and agitator speed. The current software architecture cannot ingest, normalise and contextualise these datasets across fragmented historians and the spreadsheets used during technology transfer from lab to factory.

    When instrument output lives in separate historians and laboratory notebooks, the same variable describes different things in different places; a shared ontology and a normalisation pipeline let the same query read across all of it, which is what predictive process control needs in order to work.

  • Compliance Regulatory

    Producing litigation-grade R&D provenance on demand

    The EVERY Company is defending US Patent No. 12,096,784 in litigation filed by competitor Onego Bio in the U.S. District Court for the District of Delaware. The proceeding requires absolute proof of R&D provenance, strain-engineering history and experimental outcomes with cryptographic timestamps, alongside continuous FDA GRAS compliance evidence across the full manufacturing lifecycle.

    Where records are assembled on demand for a proceeding, the time the team has to respond is set by how long it takes to gather the evidence; capturing R&D output in an integrated electronic lab notebook (ELN) and laboratory information management system (LIMS) at the moment of creation makes the same record available to a court filing and to a routine internal review.

Opportunities, by urgency and business impact

Each bubble is one opportunity, numbered to match the list below. Further right means it bites sooner; higher means a bigger effect on the business. A bigger bubble means a bigger implementation effort.

Source: A4BEE analysis of public sources

  1. Reading partner-site fermenter telemetry as a single data estate

    The R&D programme in San Francisco is digitally disconnected from the commercial fermenters operating at partner sites in Europe and Asia. Legacy SCADA and PLC stacks at the partner sites cannot publish to a cloud data lake, so ML model training and root-cause analysis of batch deviations rely on whatever has been exported manually.

    Hardened edge devices at each partner site extract the fermenter's instrument signals through OPC UA (Open Platform Communications Unified Architecture, a vendor-neutral machine-to-machine protocol) or MQTT (a lightweight publish-subscribe messaging protocol commonly used for sensor data), normalise them against a shared ontology and stream them into a single time-series data lake; a real-time dashboard with golden-batch comparison overlays shows the current run against the best historical run for the same product.

    • The EVERY Company — Company resources, 2025
    • The EVERY Company — US-based manufacturing press release, December 2024
    • AgFunderNews — EVERY raises $55M, October 2025

  2. Simulating the US and Abu Dhabi plants before construction is committed

    Two large-scale facilities are in feasibility: a US domestic plant funded through the DoD Distributed Bioindustrial Manufacturing Program, and a 4-million-liter plant in Abu Dhabi with ADIO and Vivici. Design miscalculations during feasibility translate into permanent operational bottlenecks and compressed return on investment once construction is committed.

    High-fidelity physics-based digital twins of the proposed bioreactors, downstream processing lines, utility infrastructure and material flow paths simulate fluid dynamics, oxygen mass transfer, thermal loads and mechanical shear stress; coupling the simulation outputs with a techno-economic model validates the capacity forecasts that the DoD and the ADIO investors are underwriting.

    • The EVERY Company — US-based manufacturing press release, December 2024
    • Abu Dhabi Investment Office — ADIO partners with The EVERY Company and Vivici, 2025

  3. Linking enterprise batch records across every manufacturing node

    The current batch record workflow combines paper-based and disconnected digital systems between the laboratory and the commercial production floor. For Walmart, Unilever's Vegetarian Butcher line and Grupo Palacios the contractual quality-assurance and quality-control expectations are stringent, and any traceability gap or batch-release delay is a direct contractual exposure.

    A unified manufacturing execution system (MES) connects inventory, electronic batch records and quality-control workflows across every manufacturing node, with a supply-chain visibility layer that traces each lot from upstream fermentation parameters through downstream processing and packaging to the final distribution point, exposing the same genealogy to operations and to enterprise customers.

    • AgFunderNews — McWin Capital Partners on EVERY's move from promise to proof, October 2025
    • Green Queen — Vegetarian Butcher partners with EVERY, 2024

  4. Capturing R&D provenance in an integrated ELN and LIMS

    The IP litigation with Onego Bio and the ongoing FDA GRAS compliance burden both require continuous, timestamped, cryptographically verifiable records of every strain-engineering experiment and downstream process change. The current R&D documentation spans multiple lab notebooks and standalone data files, and any fragmentation in those records is a direct legal and regulatory exposure.

    An integrated ELN and LIMS captures every R&D record at the moment of creation with an immutable audit trail, cryptographic timestamping and version control; the same record is then available for FDA GRAS submissions, for ESG lifecycle-assessment data and for the IP proceeding without a separate evidence-gathering exercise.

    • HGF — IP Ingredients: Winter Case Law Review 2025
    • AgFunderNews — EVERY and Onego Bio IP dispute, 2025

  5. Producing ESG, IP and regulatory evidence on demand

    Enterprise customers (Unilever, Walmart, Grupo Palacios) and regulatory bodies (FDA, EFSA-equivalent pathways for future market entry, U.S. DoD grant reporting) all require verified, current evidence from EVERY: ESG lifecycle assessment data for corporate sustainability reports, R&D provenance for the IP proceeding, and GRAS compliance evidence for production scale-up to new geographic nodes.

    Narrow, reviewable AI agents draft ESG lifecycle-assessment sections, assemble IP-provenance evidence packages, and produce GRAS-compliance documentation from the underlying system records; a named reviewer approves every output, and the same agent finds every controlled document a standards change touches, so the scope of a regulatory update is established by search rather than by recollection.

    • Green Queen — EVERY: Remake the Food System from Within, 2024
    • The EVERY Company — US-based manufacturing press release, December 2024

What we'd propose

  • Digital CDMO

    Unified IT/OT convergence and ML bioprocess optimisation

    Hardened edge gateways at every partner manufacturing site, a cloud data lake with a shared ontology, real-time golden-batch dashboards for the commercial fermenters, and the data pipeline and closed-loop control layer that turns EVERY's patent-pending ML bioprocess system into a live optimisation loop against the running batch.

    • Edge gateways and OPC UA extraction

      Getting fermenter signals off partner-site equipment

      Hardened edge devices at each partner site extract instrument signals (dissolved oxygen, pH, temperature, agitator speed, off-gas composition) through OPC UA (Open Platform Communications Unified Architecture) or MQTT, with encrypted transmission back to a central cloud data lake; the edge device keeps a local buffer so a connectivity drop never loses a process value.

    • Real-time dashboards with golden-batch comparison

      The current batch against the best historical batch

      A Grafana-based dashboard per fermenter shows the current run's critical parameters against the best historical run for the same product; configurable alert thresholds flag drift early enough for the operator to intervene on the running batch rather than after it has shipped.

    • Automated ML pipeline and closed-loop control

      From batch data to a control setpoint

      An automated extract-transform-load (ETL) pipeline feeds cleaned, contextualised fermentation telemetry into EVERY's patent-pending ML bioprocess system, with the model output available both as a dashboard for the process engineers and as a closed-loop setpoint signal that adjusts substrate feed rate, temperature and dissolved-oxygen targets on the running batch within a defined operating envelope.

    • R&D sees the commercial fermenter's actual operating envelope, not a reconstruction of it.
    • ML model training runs on current commercial-scale data rather than only on laboratory data.
    • Deviations surface against the batch that is running, not the batch that already shipped.

    We've done this before

    See how we do Digital CDMO
  • Digital CDMO

    Digital twin and virtual commissioning for greenfield facilities

    Physics-based digital twins of the proposed U.S. and Abu Dhabi bioreactors, downstream processing lines, utility infrastructure and material flow paths, coupled to a techno-economic model that validates the capacity forecasts being delivered to the DoD and to ADIO.

    • Bioreactor process simulation

      Fluid dynamics and thermal behaviour at commercial scale

      Physics-based models of the proposed bioreactors simulate fluid dynamics, oxygen mass transfer, mechanical shear stress and thermal loads at the planned commercial volume; the same models are run against the 5-litre benchtop recipe so the assumptions that bridge the two scales are visible rather than hidden.

    • Facility layout and utility optimisation

      Piping, utilities and material flow before construction

      A virtual replica of the facility architecture includes bioreactor banks, downstream processing lines, utility infrastructure and material flow paths, with bottleneck identification and spatial optimisation as documented outputs that go into the equipment procurement specification.

    • Techno-economic validation

      Numbers the DoD and ADIO can underwrite

      The simulation outputs feed a techno-economic model that produces the cost-of-goods-sold, energy intensity and capacity utilisation figures that appear in the DoD feasibility deliverable and in the ADIO investment case; the model is reproducible so that parameter changes can be traced through to the financial outcome.

    • Bottlenecks surface in simulation rather than after commissioning.
    • Equipment procurement decisions are made against simulated operating data, not against vendor brochures.
    • The DoD and ADIO cases are backed by the same numbers the design team is using.

    We've done this before

    See how we do Digital CDMO
  • Digital CDMO

    Integrated MES and supply-chain traceability platform

    An enterprise-grade manufacturing execution system connecting inventory, electronic batch records, quality control and supply-chain genealogy across every manufacturing node, with the data flow that Walmart, Unilever's Vegetarian Butcher line and Grupo Palacios can read directly to satisfy their own sustainability and provenance reporting.

    • Electronic batch record management

      Paperless batch documentation

      A Good-practice (GxP)-compliant electronic batch record replaces the paper and disconnected digital workflow, with enforced workflow sequencing, operator signatures, and real-time batch status across every manufacturing node.

    • Quality control integration

      QA/QC result capture at the instrument

      Analytical-instrument results feed the batch record directly, with sample scheduling, result capture and deviation management inside the same workflow; batch release waits on the analytical result rather than on the paperwork that follows it.

    • Supply-chain genealogy and ESG data feed

      Every lot traced from fermenter to customer

      A traceability layer links upstream fermentation parameters through downstream processing and packaging to the final distribution point, exposing a complete product genealogy and the underlying ESG lifecycle-assessment data to the enterprise customer through a standard application programming interface (API).

    • Batch release time depends on the analytical result, not on the paperwork that follows.
    • Walmart, Unilever and Grupo Palacios receive the same ESG and provenance data the operations team uses.
    • Manual transcription errors between laboratory, production floor and batch record are eliminated.

    We've done this before

    See how we do Digital CDMO
  • Digital Lab

    Comprehensive lab digitalisation and R&D provenance

    An integrated electronic lab notebook (ELN), laboratory information management system (LIMS) and automated workflow infrastructure that captures every R&D record at the moment of creation with an immutable audit trail, providing the evidence base for the IP litigation, the FDA GRAS submissions and the technology transfer from San Francisco to partner manufacturing sites.

    • GxP-compliant ELN with cryptographic timestamping

      R&D records that hold up in court

      An electronic lab notebook with immutable audit trails, cryptographic timestamping and version-controlled strain-engineering records captures every experiment at the moment of creation; the same record is then available for FDA GRAS submissions, for the IP proceeding and for ESG lifecycle-assessment data without a separate evidence-gathering exercise.

    • Unified LIMS and ELN data architecture

      One laboratory data layer

      A unified data layer connects ELN experimental records with LIMS sample management, analytical results and master cell bank registries; sample genealogy flows from the benchtop strain-engineering record through analytical characterisation and into the technology-transfer package that travels with each strain to the partner manufacturing site.

    • Automated regulatory submission packages

      GRAS, IP and ESG evidence assembled from the record

      Reporting workflows generate FDA GRAS compliance documentation, IP-provenance evidence packages and ESG lifecycle-assessment data exports directly from the integrated lab data architecture, with each output traceable to the underlying records that produced it.

    • Every R&D record is captured once and reused across GRAS, IP and ESG submissions.
    • The technology-transfer package to a new partner manufacturing site is generated, not assembled by hand.
    • The IP litigation evidence chain stands on the underlying records rather than on reconstruction.

    We've done this before

    See how we do Digital Lab
  • Agents

    AI agents for ESG, IP and regulatory documentation

    Narrow, reviewable AI agents that draft ESG lifecycle-assessment sections, assemble IP-provenance evidence packages and produce GRAS-compliance documentation from the underlying system records; a named reviewer approves every output, and the same agent finds every controlled document a standards change touches.

    • Drafting from system records

      First drafts from real data

      Agents draft the first version of ESG lifecycle-assessment sections, IP-provenance evidence summaries and GRAS-compliance documents directly from the ELN, LIMS, MES and cloud data lake records, so the human author edits and judges rather than assembles.

    • Template and completeness checking

      Gaps found before review

      Before any document enters the human review queue, an agent checks it against its template (FDA GRAS, ESG reporting standard, court-required evidence format) and against the site's own checklist, returning the missing or inconsistent sections so they are filled before a reviewer opens the file.

    • Change-impact search across the controlled document set

      Which documents a change touches

      When a regulatory standard changes, the agent retrieves every controlled document that references it and ranks them by how directly they are affected, so the scope of the update is established by search rather than by recollection; the same capability finds the records the IP proceeding's discovery requests name.

    • Submission packages arrive at review complete rather than partial.
    • The scope of a standards change is established by search rather than by recollection.
    • Every output is traceable to the source records it came from and signed off by a named reviewer.

    We've done this before

    See how we do Agents

Where QB Systems fits

Alongside our services we build QB Systems, hardware and software for bioprocess control. QB Systems is a product brand of A4BEE Sp. z o.o.

Applications
Bioreactors

Software-defined control for a bioreactor — a new QB vessel, an upgrade to one you have, or a retrofit of the existing PLC.

Automated sampling

Automated sampling from 4–18 sources, aseptic-capable and up to 72 hours unattended. Works with any vendor's bioreactor.

Buffer & media preparation

Automated preparation of growth media and process buffers, so a recipe runs the same way every time without fixed infrastructure.

Deployment
Hybrid connectivity

QB talks to equipment already in place over OPC-UA or Modbus, leaving the vendor's own control in charge.

Scale
Benchtop (1–8 L)

Glass vessels with the complete hardware and software stack. This is the core range for development work.

Digital maturity: today and target

Scored out of 100 across six dimensions. The target is what The EVERY Company's own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
IT/OT convergence 25 → 85
R&D in San Francisco and commercial production at partner sites in Europe and Asia run on separate data estates; partner-site SCADA and PLC stacks were not designed for secure cloud connectivity, so ML model training currently relies on manually exported data rather than live fermenter telemetry.
Data architecture 30 → 90
Benchtop and commercial fermenters generate continuous time-series data on dissolved oxygen, pH, temperature and off-gas composition, but the current architecture does not ingest, normalise and contextualise these datasets across fragmented historians and the spreadsheets used during technology transfer.
Manufacturing execution 20 → 80
Batch records currently combine paper-based and disconnected digital systems between the laboratory and the commercial production floor; the enterprise customer base (Walmart, Unilever, Grupo Palacios) requires integrated batch genealogy and quality control as a contractual expectation rather than a stretch goal.
Lab digitalisation 35 → 85
R&D generates continuous genomic and phenotypic data, but the documentation is split across multiple lab notebooks and standalone data files; the IP proceeding with Onego Bio and ongoing FDA GRAS compliance both depend on the same underlying R&D record being assembled on demand from a single source.
Greenfield simulation 10 → 75
The US domestic facility under the DoD Distributed Bioindustrial Manufacturing Program and the 4-million-liter ADIO/Vivici plant in Abu Dhabi are both at feasibility stage; no virtual facility models exist yet for either site, so scale-up decisions currently rest on engineering judgement and vendor specifications.
Cybersecurity and compliance 30 → 80
Distributed manufacturing across third-party partners expands the operational technology (OT) attack surface; legacy OT systems at partner sites lack modern security protocols, and ESG data verification infrastructure is still being assembled for enterprise sustainability reporting requirements.

Check this yourself

Our Service Portal has free self-assessments and market comparisons. These are the ones that line up with what we've read above — no sales call required.

Think we've read this right?

Talk to us

Related reading

This is an independent analysis prepared by A4BEE from publicly available information as of March 2026. It reflects A4BEE's own interpretation and opinion, is not affiliated with, endorsed by, or verified with The EVERY Company, and may be incomplete or inaccurate. All company names and trademarks are the property of their respective owners. To request a correction or removal, contact [email protected].