Working at Manifold Bio: Culture, Pace and Who Thrives

How Work Gets Done at Manifold Bio

By the end of 2021, Manifold Bio's platform tracked over 150 mCode barcodes, a hundredfold jump over the industry standard for multiplexed in vivo screening, BioCentury reported. That number is the lead: a single mouse can now carry a library of protein variants large enough to turn animal testing from a bottleneck into a data engine.

Most biotechs separate the people who design molecules from the people who test them in animals. Manifold puts them at the same bench. The company runs a single interdisciplinary loop: computational scientists design proteins, wet-lab teams run in vivo screens the same week, and in-house automation feeds data back to the models. That loop — design, test, learn, repeat — drives hiring, team structure, and decision rights. Throughput sets the pace. In the 2023–2024 push toward a blood-brain barrier shuttle, the team iterated through tens of thousands of designs, each tested in a living animal.

The culture rewards speed and ownership while creating burnout risk for people who need structure. Decisions are made close to the bench by small interdisciplinary teams. Authority follows the experiment. When Shane Lofgren sent a 2,000-word email dismantling the early commercial strategy, Gleb Kuznetsov asked for his alternative and got another 2,000 words. The board backed Lofgren's bet on the blood-brain barrier shuttle, and the company narrowed its platform scope to execute it.

Chief technology officer Pierce Ogden built the automated rig that executes complex experimental workflows at the push of a button. The rig sits inside a Boston facility designed to force interaction: open work areas, meeting rooms, and breakout zones are placed throughout to promote collaboration among the multidisciplinary team. The org chart shows 13 executives, but working units stay small. A Director, Oligo Team Lead runs oligonucleotide synthesis. A Computational Scientist, Assay Development owns the analysis pipeline. A Product Manager, AI Platform & Partnerships translates platform capability into partner-facing product.

That decision-making style — fast, founder-proximate, data-settled — extends to business development. The Sr. Director / VP, Business Development does not shop services contracts. Manifold's model, Kuznetsov said, is "we don't sell labor; we sell rights to certain applications of our intellectual property." Deals transact on cells in the array: pharma partners license the shuttle for a specific payload and indication. Steve Holtzman, hired as Executive Chairman, structures those agreements. The Roche partnership, announced October 2025, emerged from years of technical de-risking facilitated by Nucleate, not a cold pitch.

Contract roles exist at the bench; a (Contract) Research Technician, Pre-clinical Pharmacology earns an hourly rate, but the salaried band reflects a team built for ownership, not task execution. The median salaried compensation sits at a competitive level across eight roles. The loop only works if the person running the in vivo study can call the computational scientist and change the next design batch. That proximity is the culture.

Values That Drive the Machine

Manifold Bio's operating philosophy crystallized early. In a 2022 interview, Kuznetsov named two explicit values: scientific boldness and team focus. A third principle — ownership — runs beneath both. "One is having a principle of ownership: very clear, who owns what, who's responsible for something," he said. "We make sure it's very clear who owns what, but then share in both the success as well as failures." That formulation — individual accountability paired with collective upside — shapes how the company assigns work and evaluates outcomes.

The "town square" ritual embodies that culture. Kuznetsov described a scientific culture with a "town square around experimental design, where the team comes together to design experiments that produce the right data to enable quick decision-making and iteration." The phrase signals that experimental design is not a solo act; it is a public, cross-functional negotiation. Data generation serves decision velocity, not publication volume. The same logic drives hiring: from the start, the company used case studies where candidates work through a problem with the team. "One very effective thing we've done from the beginning is a case study, where we work through a problem together," Kuznetsov said. The format tests technical judgment and collaboration simultaneously, with no separate culture interview required.

Team focus is not a soft-value add-on. Kuznetsov framed it as the primary draw for academics leaving the lab: "A common pattern we see for the reason people transition from academia to industry... is because they're looking for more of a team-based approach than academia often offers." He made the stakes explicit: "You're working together to accomplish a big, complex problem. That's the core, and the kinds of people that have joined the team need to care about that: how do we do something big together and combine skills and combine expertise?"

That expectation translates into a proactive ownership model. Describing the Church lab environment that shaped him, Kuznetsov said in 2026: "It creates this selection pressure for people who know what they want to do and how to ask for help. It creates this training ground for future professors and entrepreneurs, where you learn to be very proactive." George Church's mentorship style reinforced this: "You're the captain of your ship. What do you think you should do?" The metaphor — captain, not crew — implies latitude to set direction, but also responsibility for the vessel.

A related mental model: the 1% threshold. Church would say: "1% is tantalizingly close to 100%. That was a very powerful idea. If you can get from zero to one, then you've shown that it's possible, and now it's an engineering problem." The line reframes early proof-of-concept not as a milestone but as a phase change: once existence is proven, the work becomes systematic. "Think of it as an opportunity, not an obligation," Church added, a framing that recurs in how the company talks about scale.

Platform thinking operates at the strategy layer. Kuznetsov distinguishes lowercase-p platforms — a single architecture like a T-cell engager, from uppercase-P platforms with "a prolific space." "The platformness of a company is defined by the breadth of the product space. How many products can you really create?" That breadth determines commercial strategy: how much partnering versus proprietary asset development the business can sustain. Manifold's commercial strategy explicitly spreads risk across partnerships and proprietary programs rather than betting everything on a single drug candidate, a discipline Kuznetsov traced to a 2023 board conversation. Facing a shrinking funding market and rising oncology data bars, the company pivoted to brain delivery: "we should really double down in this brain direction... It turned out to be a very good decision."

The data moat is the operational expression of platform breadth. "It's partly a statement about what we think is important, which is the data moat," Kuznetsov said. "We're doing some of the most innovative AI work, and maybe folks didn't fully appreciate that we're not just doing in vivo multiplexing." The moat requires domain depth: "Getting deeply expert in the domain is the key to inventing the key enabling piece that takes you from uncanny valley to really useful." He warned against "throwing AI at the wall": "when you use AI on something that you're deeply expert in, there's this constant uncanny valley experience because you can see the flaws." The remedy: "Take seriously getting the domain knowledge of whatever problem you want to get into."

Curiosity acts as the self-selecting filter. "We just look for curiosity about the biology. It's almost self-selective," Kuznetsov said. The company cannot outbid OpenAI on salary, but it offers something narrower: "Knowing you're accessing data that no one else has is also very addicting." That data advantage, hundreds of thousands of designs tested in a single animal, is the foundation of the AI feedback loop. "You have to have an advantage of the data to guide the AI. And be able to test the outputs of the AI."

The mission statement on the company site makes the endpoint explicit: "We're developing foundational technology that will refactor the entire biologics drug discovery and development process and contribute meaningfully to improving human health." The long-horizon vision, "Towards the virtual organism: In the limit, we're building a predictive model of living systems, one that integrates high-throughput in vivo data to map how biomolecules behave in the complexity of the body," frames every near-term decision as a step toward that model.

Founder behavior reinforces the values. The early days, when Kuznetsov, Ogden, Eric Kelsic, and Surge Biswas constantly helped each other debug experiments, brainstorm, and try new things, established a norm of peer-level technical collaboration. Lofgren's arrival as the first person "really thinking deeply about our product" shifted the company from technology-first to product-first: "Before Shane, our product applications were really underbaked. Like, what was the connection between our platform and product?" The Roche relationship, "Literally, we've been talking to Roche since we started the company," demonstrates the long-horizon partnership discipline that the uppercase-P platform strategy demands.

Together, these principles form a coherent system: bold scientific bets, owned by individuals, debated in the open, executed by interdisciplinary teams, measured against a data moat that compounds with each experiment. The culture rewards the person who can work through that loop without a map. It penalizes the one who waits for a map to be handed down.

What the Hiring Filter Catches

The acceptance rate sits at 3 to 5 percent for qualified candidates, per interview guides published in March 2026. That figure alone tells you the filter is tight. But the more useful signal is what the filter actually catches: people who can operate at the intersection of wet-lab biology and computational scale without waiting for a handoff.

Technical depth is the price of entry. Data engineering roles demand advanced Python and SQL (specifically PostgreSQL) plus production-grade ETL pipeline design, data modeling for evolving scientific datasets, and AWS fluency. Benchling experience is a plus. Research associates need molecular biology, mammalian cell culture, and screening assay experience. Computational scientist and AI/ML research engineer postings ask for library-guided design and hybrid computational/wet-lab expertise.

The roles that sit at that intersection carry the widest bands. The table below summarizes the salaried and contract roles with their locations and compensation bands, Zero G Talent's job board data shows.

Median across eight salaried roles: $182K.

Domain alignment is the next filter. Candidates must demonstrate they understand Manifold's core platform: massively parallel in vivo screening via protein barcoding, where up to a hundred simultaneous tests run in a single mouse. CTO Pierce Ogden has said the edge isn't AI itself; it's pairing AI with the quality, relevant data that platform produces. Interview guides explicitly tell candidates to prepare to discuss how scalable data engineering accelerates targeted biologic discovery. That means reading the TechCrunch coverage, the BioPharma Dive pieces, and the company's own site before the first conversation.

The third signal is bridge-building. Every technical role description emphasizes collaboration with both computational and wet-lab scientists. The behavioral screen probes for past work translating scientific requirements into technical solutions and communicating complex data concepts to non-technical stakeholders. This isn't a courtesy skill. In a company where decisions happen at the bench, the person who can explain a pipeline failure to a biologist and propose a fix before the next run owns the outcome.

Data integrity is the quiet disqualifier. Guides stress validation, documentation, feedback loops, and reproducibility. Manifold's platform generates the training sets for its own models; garbage in means garbage compounds. Candidates who treat data quality as a compliance checkbox don't last.

The company also states a commitment to cultivating a diverse workforce, noting that varied perspectives drive impactful healthcare solutions. That language appears in public hiring materials, but the lived test is whether a candidate can operate in a small interdisciplinary team where structure is minimal and ownership is expected. The hiring bar selects for people who have already done that and want to do it again.

The Review Gap

Public employee reviews for "Manifold" on Indeed show a 4.4 overall rating across five reviews spanning 2013 to 2019. Category scores range from 3.8 to 4.5. But these reviews describe an "award-winning experiential agency" with "events" staffing issues, "shoveling snow" as the hardest part of the job, and independent contractor arrangements, language that matches a marketing or events business, not the biotech company Manifold Bio that designs tissue-targeting medicines and runs computational protein engineering teams.

The disconnect is material. The Indeed reviews reference "Manifold" and "Manifold Solutions," not "Manifold Bio," and the described work (event staffing, consumer-facing contract work, snow removal) bears no resemblance to the roles on Manifold Bio's current hiring board.

The board lists eight salaried roles with a competitive median band and founders Shane Lofgren, Gleb Kuznetsov, and Pierce Ogden, a founding team building a platform that, per Anthropic's 2026 reporting, uses Claude Science to nominate targets by assessing surface expression, trafficking, and safety against proprietary internal data.

No Glassdoor, Levels.fyi, or Blind reviews specific to Manifold Bio surfaced in the research. The only dated, attributable employee commentary for the biotech entity is the Anthropic case study, which quotes Manifold saying Claude Science "could do this end-to-end, gathering the right data and applying the right judgment with the context of past programs built in," a statement about tooling, not culture. The Indeed reviews, while verifiable as of their posted dates (2013-12-14 through 2019-11-21), almost certainly reflect a different employer sharing the "Manifold" name.

This gap matters for candidates. A biotech running that product manager role and Director, Oligo Team Lead operates in a different labor market, with different pressures, than an experiential agency. The absence of attributed, recent employee accounts from the actual Manifold Bio means any culture assessment must rely on the company's operating model (small interdisciplinary teams, decisions close to the bench, founder-led) rather than verified employee testimony. Candidates should probe directly on autonomy, pacing, and burnout risk during conversations with the team.

Who Thrives, Who Burns Out

The candidate who thrives at Manifold Bio looks less like a traditional biotech hire and more like a founder's first employee. The company's 63.5 percent headcount growth over the past year, its founder-led structure (Lofgren, Kuznetsov, Ogden), and its reliance on small interdisciplinary teams that decide close to the bench all select for people who treat ambiguity as a design space rather than a defect.

Autonomy is the baseline. A Scientist, cDNA Display and an Associate Scientist in In Vivo Pharmacology both sit on small teams where experimental calls happen without a committee. That product manager sits at the same intersection, translating internal capability into external deals, a role that only works if the PM can speak the science fluently.

Burnout risk concentrates where autonomy meets pace. That business development lead and Director, Oligo Team Lead carry significant accountability in a 51–200 person organization that is still building its process layer. The "virtual organism" ambition Kuznetsov described means the scientific surface area expands rapidly. Those contract technicians see the same tempo at the bench: high-throughput in vivo screens generate large data volumes per run, and the turnaround from readout to next design cycle is fast.

People who need explicit prioritization, defined career ladders, or a manager who shields them from strategic churn will struggle. The office design (break areas, recreational spaces, sustainability features) signals intent around recovery, but it cannot substitute for structural slack that does not yet exist. Candidates who have only operated in process-heavy CROs or large pharma should expect a culture shock: the first month is contributing to a live program. The ones who stay are the ones who would have built the tool they needed rather than filing a ticket for it, the same ones who, handed a blank bench, start running the loop before anyone draws the org chart.

Working in frontier tech? Zero G Talent tracks the openings: see every open Manifold Bio role, browse frontier tech jobs, the companies hiring, and the people building the field.