How work actually gets done
The blood draw is simple. Everything behind it is not.
Freenome operates at the intersection of two clocks that refuse to synchronize. One ticks to the rhythm of biology: patient enrollment, sample collection, longitudinal follow-up, the slow accumulation of clinical evidence that regulators demand. The other ticks to the speed of compute: model iteration, GPU allocation, the weekly cadence of deep-learning experiments that can rewrite a classifier's performance overnight. The daily work is the friction between them.
Roughly 410 people work out of the Brisbane, California headquarters, with a sizable remote contingent across computational and data roles, per LinkedIn data. The organizational structure runs through three pillars: the wet lab and clinical operations that generate multiomic data, including methylation, RNA, protein, metabolomics from a single blood draw; the machine-learning and computational-biology teams that turn billions of data points per sample into classifiers; and the regulatory, commercial, and health-system functions that translate a cleared test into a reimbursed clinical pathway.
Decision-making is gated by the strictest clock in the room: the FDA. The PREEMPT CRC study that underpinned the SimpleScreen™ CRC approval enrolled nearly 50,000 average-risk adults and published in JAMA in June 2025; the Premarket Approval Application received FDA approval in July 2026. That timeline — years of prospective collection, a locked statistical analysis plan — forces a discipline that pure software shops rarely face. The company's public filings list "regulatory approval timelines for diagnostic tests" and "clinical validation of product candidates" as top risks — a reminder that the critical path runs through biology, not code.
The partnership structure adds another layer of governance. Exact Sciences holds exclusive U.S. commercialization rights for the colorectal test; Abbott will exclusively commercialize the lung test in the U.S. starting fall 2026; Roche holds ex-U.S. rights to develop "kitted" versions that decentralize processing. The Roche deal (more than $200 million including a $75 million equity instrument, milestones, and royalties, as reported in Freenome's November 2025 PR Newswire announcement) and the roughly $300 million in gross proceeds from the July 2026 public listing (ticker: FRNM) mean capital is less constrained than at most private biotechs, but allocation decisions now face public-market scrutiny.
Compensation reflects the dual-discipline reality. First-party board data from Zero G Talent shows salary bands well above typical biotech medians:
| Role | Salary Band |
|---|---|
| Vice President, Product Management | $294k–$384k |
| Senior Director, Real World Data | $236k–$335k |
| Staff Machine Learning Scientist | $200k–$284k |
| Staff Computational Biologist | $188k–$270k |
| Overall median (23 roles) | $227k |
Equity is a meaningful component; the company raised over $1 billion before going public, and the PIPE led by Perceptive Advisors and RA Capital included participation from Bain Capital Life Sciences, Farallon, and ADAR1, but the transition to Nasdaq introduces quarterly reporting, lock-up schedules, and a visible share price that changes how employees value their grants.
The lab received CLIA certification in October 2025, enabling the early-access program that shipped first patient samples in January 2026. The Vallania and Sanderson multi-cancer studies are fully enrolled, feeding over ten additional indications into the same assay. The PROACT Lung study is ongoing. Each is a hard deadline the AI side must plan around.
In this cohort, Freenome's proprietary multiomics platform -- using base-resolution methylation sequencing of circulating cell-free DNA (cfDNA) and plasma protein immunoassays -- achieved adjusted sensitivity of 90.7% at 50% specificity and 80.4% at 75% specificity for detecting lung cancer, according to Freenome's March 2026 PR Newswire release. The multiomics test detected lung cancer across all three subtypes evaluated in this study (adenocarcinoma, squamous cell carcinoma and small-cell lung cancer) and across all disease stages, including an adjusted sensitivity of 77.1% at 50% specificity for Stage I cases.
"Less than 20% of eligible high-risk adults are currently screened for lung cancer, the leading cause of cancer death in the U.S.," said Jimmy Lin, M.D., Ph.D. MHS, chief scientific officer at Freenome. "A blood-based screening test could provide a more accessible option to increase screening participation. In this study, our multiomics approach demonstrated the potential of our test to detect lung cancer across stages and subtypes, and we're encouraged as we continue to advance to a larger validation study in a previously unseen evaluation cohort."
Freenome is developing a flexible multi-cancer detection platform designed to support a personalized test offering tailored to each individual's health status, risk factors and screening recommendations. Leveraging a single blood draw and a common assay, this approach utilizes machine and deep learning classifiers to optimize diagnostic accuracy across a diverse range of cancer types.
The values that govern the work
Freenome lists five values on its website — Strive for Greatness, Servant Leadership, Trust, Integrity, Empathy — and says they guide everything the company does. In biotech, where the gap between a press release and a patient result spans years, that claim gets tested daily.
Strive for Greatness frames the difficulty as the point: "We choose to take on cancer not because it's easy, but because it's hard." The technical bet is that no single analyte class suffices; the advantage comes from "learning across biological signals, products and cancer types, not from a single static test," as co-founder and chief product officer Riley Ennis said in July 2026. That philosophy drove the nearly 50,000-patient study published in JAMA. The results (roughly four in five cancers detected at nine in ten specificity) became the basis for FDA approval of the SimpleScreen CRC test.
Servant Leadership ("We lead by example and help others grow to their full potential") shapes organizational choices. Freenome's research programs span discovery, development, and validation. Its commercial strategy leans on partnerships with Abbott and Roche rather than building a solo sales force. The "Freenomer" definition the company publishes ("a mission-driven employee who is fueled by the opportunity to make a positive impact on patients' lives, who thrives in a culture of respect and cross collaboration") makes cross-disciplinary work a cultural expectation. When the company describes its three foundational pillars (Building a Comprehensive Cancer Detection Portfolio, Advancing the AI-Enabled Multiomics Platform, Scaling Commercialization), each pillar demands that biologists, computational scientists, and commercial operators share roadmaps.
Trust, defined as valuing "long-term relationships, not short-term transactions," shows up in the cap table and the partner list. A syndicate built for decade horizons includes Perceptive Advisors, RA Capital Management, Roche Venture Fund, Kaiser Permanente, Novartis, and the American Cancer Society's BrightEdge Ventures. The July 2026 business combination with Perceptive Capital Solutions Corp. brought that amount to fund the next phase. Abbott and Roche partnerships, announced alongside the public debut, cover commercial, laboratory, and digital infrastructure for delivering multiple screening tests across health systems.
Integrity ("We courageously do what is right, and are fact-based, data-driven, and accountable for results") is visible in the regulatory record. The JAMA publication, the FDA approval that followed, the Breakthrough Device Designation granted to SimpleScreen Lung — each is a public, auditable milestone. The company's description of its data flywheel ("As more patients are tested, the resulting data creates a data flywheel that continuously improves test performance and accelerates future development") only holds if the initial data meets a standard high enough to earn clinical adoption. The PROACT Lung study, underway as of mid-2026, applies the same evidentiary bar to a second indication.
Empathy ("We seek to understand before being understood and build products that equitably serve diverse communities") addresses the market reality that roughly 60 million eligible Americans are not up to date on colorectal screening. A blood draw lowers the barrier compared to colonoscopy or stool tests. The Personalized Cancer Detection framing ("screening as unique as each individual patient") ties the value to product architecture: repeated screening establishes a personal baseline, and longitudinal clinical outcomes feed back into risk models.
Together, the five values function as a filter for the three strategic pillars. Building a portfolio across over ten additional indications requires Strive for Greatness. Advancing the AI-enabled platform demands Integrity in model validation and Servant Leadership between wet-lab and dry-lab teams. Scaling commercialization through health systems runs on Trust with partners and Empathy for the unscreened population. The values are not the culture; they are the terms the culture agrees to be judged by.
What the hiring bar selects for
Freenome's interview process filters for people who can operate at the intersection of wet-lab biology, large-scale machine learning, and regulated product development — a combination that rules out most single-discipline specialists. The company's own definition of a "Freenomer" makes the bar explicit: a mission-driven employee fueled by patient impact, who thrives in cross-functional collaboration, and whose work moves both the company and their own career forward. That phrasing maps directly to the roles the company has been filling. Board data shows 23 salaried openings with a median band around $227,000, spanning roles that only exist where biology, AI, and clinical translation meet.
Mission alignment is tested behaviorally. In a 2022 interview, then-CEO Mike Nolan reiterated the same rationale, and the hiring loop probes whether a candidate has persisted through long, ambiguous scientific programs. The registrational study for the colorectal test enrolled tens of thousands of subjects with matched colonoscopy and histopathology reports. Candidates who have shipped a diagnostic from discovery through pivotal trial to CMS coverage discussion (the roughly 74% sensitivity / 90% specificity threshold that triggers national coverage) carry a signal that a pure publishing record does not.
Diversity of data (and of the teams that build on it) is another explicit filter. Nolan noted that Freenome deliberately matched its registrational cohort to U.S. Census figures for clinical, socioeconomic, and geographic representation. The public listing, backed by a roughly $300 million raise, has sharpened the bar for execution speed without sacrificing rigor. Roles now carry explicit delivery milestones: the PROACT Lung study readout, next-generation assay automation, and the infrastructure to deploy multiple tests across primary care.
In practice, the hiring bar selects for three intersecting competencies: deep technical credibility in at least one of the core disciplines (multiomics, ML, clinical development, commercial translation), demonstrated ability to collaborate across the other two, and a track record of finishing hard, regulated projects that affect patients. The salary bands reflect that scarcity: staff roles range from roughly $190k to $380k. The company does not hire for potential alone; it hires for proven ability to move a multiomic-AI diagnostic from bench to bedside.
The view from inside
Public employee-review data for Freenome is sparse. The research for this profile did not surface any attributed, on-the-record quotes from current or former Freenome staff discussing day-to-day culture, management style, or work-life balance.
The board data shows a hiring pattern skewed senior: director-and-above titles dominate, most remote-eligible, with that median across 23 roles. That composition aligns with the company's stated model, blending rigorous biological research with rapid AI development, which demands deep domain expertise on both sides. But without employee narratives, it is impossible to verify whether the cross-disciplinary collaboration the company emphasizes translates into daily practice, or whether the "high bar" referenced in internal messaging manifests as mentorship or as pressure. The board data confirms compensation is competitive for senior technical and commercial roles; it does not speak to equity refresh cadence, promotion velocity, or how the organization handles the inevitable friction between wet-lab timelines and model-iteration cycles.
In the absence of review volume, the most reliable proxy remains the hiring signal itself: Freenome is recruiting for ownership-level roles at above-market cash bands, and it is doing so remotely for many of them. That implies a degree of autonomy and trust in senior contributors. Whether that autonomy extends to mid-level engineers and scientists (and whether the culture sustains the "rigorous biological research" side as aggressively as the "rapid AI development" side) cannot be answered from the public record today.
Who stays, who leaves
The Freenome employee profile the company describes ("mission-driven," that same drive, thriving in that same culture) maps directly onto the operational reality of a 200-to-500-person organization that just simultaneously secured FDA approval for its first blood test, went public on Nasdaq, and announced updated clinical data showing improved precancerous lesion detection. That simultaneous FDA approval, public listing, and data readout wasn't an anomaly; it was the compression of a decade-long bet that multiomics plus machine learning could crack early cancer detection. People who stay tend to treat that bet as a personal stake, not a job description.
The cross-disciplinary architecture is the first filter. Freenome's platform sits at the intersection of molecular biology (cell-free DNA methylation at single-base resolution), computational biology, machine learning, clinical trial execution (the nearly 50,000-patient PREEMPT CRC study), regulatory strategy (Breakthrough Device Designation for the lung test), and commercial deployment through the Abbott and Roche partnerships. A staff computational biologist and a VP of product management are expected to speak each other's languages daily.
The pace filter is real and documented. Leadership describes the team as "relentless in pushing for what's possible." The July 2026 milestone cluster — FDA approval, Nasdaq listing, updated CRC data readout, team celebration at Brisbane HQ — followed years of clinical validation work that produced the JAMA-published PREEMPT CRC results (about 83% cancer detection, nine in ten specificity). Now the company is scaling commercialization infrastructure, running the PROACT Lung study, and advancing over ten additional indications. That trajectory means periods of sustained intensity around regulatory submissions, data locks, and partner integrations.
The mission filter cuts both ways. "Outpacing cancer starts with early detection" is a recruiting magnet for people who lost relatives to late-stage diagnosis or who came from pure research and want translational impact. The integrity value ("fact-based, data-driven, accountable for results") means negative data gets surfaced fast. A failed assay or a model regression isn't a career event; hiding it is.
The trust value (that principle) shapes tenure expectations. With over $1 billion raised from investors including Roche, Novartis, Kaiser, and that organization's BrightEdge Ventures, and a board featuring former Roche Diagnostics head Ann Costello and former NeoGenomics CEO Douglas VanOort, the company operates on horizons measured in clinical-trial years, not funding-round quarters.
Remote work adds another dimension. Senior roles in real-world evidence, evidence generation, machine learning, and computational biology are listed as remote alongside Brisbane-based leadership roles. That distribution works for people who default to asynchronous, documentation-heavy collaboration. The "cross collaboration" the company prizes requires deliberate over-communication when half the computational team is distributed.
The company's next phase (scaling SimpleScreen CRC through primary care workflows, advancing SimpleScreen Lung through PROACT, building the data flywheel from population-scale testing) will select for the same profile again. The blood draw is still simple. The work behind it never will be.
Working in frontier tech? Zero G Talent tracks the openings: see every open Freenome role, browse frontier tech jobs, the companies hiring, and the people building the field.