Careers at SentiLink: Teams, Pay and How to Get Hired

The hiring picture: two modes, one board

SentiLink's public job board shows seven open roles, all in sales, topping out at $550,000; Zero G Talent's board data shows a salary band of $120k–$350k with a median of $240,000 across 37 salaried roles. SentiLink's site reports that The company verifies three million identities a day for 13 of the 15 largest U.S. banks, six of the top 10 credit unions, and 50-plus fintech unicorns. That volume doesn't run on sales alone. It runs on the risk analysts who catch the fraud patterns the models miss, the engineers who ship API responses in milliseconds, the data scientists who retrain synthetic-fraud scores against shifting attack vectors, and researchers like Dr. David Maimon, Head of Fraud Insights, who has briefed Congress on government-program fraud and published whitepapers on nation-state threat actors.

The seven live postings cluster entirely in commercial functions: Head of Strategic Sales ($450,000–$550,000), Head of Crypto GTM ($275,000–$350,000), Enterprise Account Executive for Insurance ($275,000–$350,000), and four Account Executive slots split across state government, SLED East, and SLED West (each $300,000–$350,000). That concentration reflects a deliberate push: Forbes found that the company added 50 customers in 2024 and doubled daily identity checks in the first two months of 2025. Scaling revenue to match that volume requires a sales engine built for enterprise procurement cycles; hence the specialized AE roles for insurance, crypto, and public-sector verticals.

But the commercial push sits on a technical foundation that predates it. Co-founders Naftali Harris and Maxwell Blumenfeld met at the University of Chicago and cut their teeth at Affirm, where Harris became the first data scientist and later led the Risk Decisioning team while Blumenfeld ran Risk Operations. Their founding insight — twelve loan applications, same name and date of birth, twelve different SSNs, all with 700-plus credit scores — came from manual case review, not a model. That DNA persists. The company still runs a weekly all-hands case review every Thursday, and its Risk Operations team manually inspects a subset of daily applications to surface new fraud vectors before they scale. The product suite, including Fraud Scores, CIP Match & Watchlists, Intercept, Facets, eCBSV, and Insights, each demands distinct technical ownership: low-latency API infrastructure, feature engineering for identity attributes, compliance-grade watchlist matching, real-time Social Security verification via the Social Security Administration's eCBSV service (SentiLink was the first provider to offer it), and investigative tooling for analyst workflows.

The hiring mix, then, is bimodal. One mode is visible on the board right now: senior commercial operators who can navigate multi-stakeholder deals in regulated verticals. The other is the steady, less-publicized recruitment of engineers, data scientists, risk analysts, and fraud researchers who keep the detection layer ahead of the adversaries. The board's salary band ($120,000 to $350,000) spans both modes. A Staff Engineer and a Strategic AE sit at the same ceiling. The median of $240,000 suggests the bulk of roles cluster in the senior individual-contributor range across functions. For a candidate, the question isn't whether SentiLink hires your profile. It's which hiring cycle you're catching.

Compensation: what the numbers show

SentiLink's compensation structure reflects a company that learned from early startup experience. When the company launched in 2017, the founders set their own salaries at a nominal level, a symbolic commitment to capital efficiency that contrasts sharply with the board data today.

BuiltIn's employer-verified summary confirms defined commission tiers and company equity across the board. For non-sales roles, the median $240k base aligns with a 71-person product and engineering organization that competes for talent against fintechs and major banks.

Equity follows a standard Silicon Valley template: four-year vesting with a one-year cliff, a structure the 2019 talk explicitly endorsed. The same talk stressed the importance of filing an 83(b) election for any vested stock grants, a detail that matters for early joiners. The company provides relocation assistance and a home-office stipend for remote employees.

Benefits are unusually comprehensive. Health coverage is 100 percent employer-paid for employee and dependents across medical, dental, and vision. Parental leave runs 14 weeks for birthing parents and 10 weeks for non-birthing parents, fully paid. The 401(k) includes a company match. Flexible time off, wellness days, a birthday day off, and paid holidays round out the time-off package. Fertility benefits, dependent care FSAs, mental health coverage, and disability and life insurance (including supplemental) fill out the safety net.

The compensation philosophy articulated in 2019 still echoes: "Come up with salaries that you can realistically afford as a company." Today, with three million identities verified daily and a customer base that includes the largest U.S. financial institutions, SentiLink can afford a structure that sits above market median for technical roles while using variable pay to align sales incentives with the fraud-detection revenue engine.

The interview gauntlet: what actually gets tested

SentiLink's interview process carries a 5.4 out of 10 difficulty rating across 47 candidate reports on Dataford, placing it squarely in "medium" territory: harder than a checkbox screen, lighter than a FAANG-style gauntlet. Glassdoor mirrors that volume with 48 posted questions and reviews, and Dataford maintains dedicated guides for seven roles: Account Executive, Backend Engineer, Data Engineer, Data Scientist, Product Manager, Software Engineer, and Strategy & Data Analyst. The topics tested shift sharply by function. Sales-facing tracks lean on GTM strategy, pipeline management, and communication structure. Engineering and data tracks hit Python, system design, data pipelines, platform architecture, and fraud-specific analytics: credit modeling, fraud systems, and the data infrastructure that powers them.

The first filter is typically a recruiter screen. Candidates report consistent baseline questions: walk me through your background, why this role, why SentiLink. The recruiter is moving fast, with industry patterns suggesting 10 to 20 conversations a day, so concise, rehearsed answers that hit the mission alignment signal (fraud prevention, identity verification, the economics of synthetic identity) move you forward. From there the branches diverge. Account Executives typically face a sales manager deep-dive, a live exercise (cold call or discovery mock), and a final culture check with a VP or CRO. Data and engineering candidates move into technical screens: SQL and Python work, pipeline design, a system-design session, and a domain conversation around fraud signals or credit-risk modeling. Dataford's topic weights show fraud systems and credit analytics each appearing in 96% of relevant guides; data pipelines hit 94%; system design 96%.

What separates advances from rejections? The public reviews cluster around three signals. First, specificity on fraud domain knowledge: candidates who can discuss synthetic-identity clusters, velocity rules, or the trade-offs between false positives and catch rates score higher than those treating it as generic ML. Second, structured communication: the "interview process communication" tag appears in 100% of Dataford's SentiLink guides, and reviewers repeatedly note that disorganized answers to "tell me about yourself" or "why SentiLink" stall candidates before the technical rounds. Third, evidence of ownership: not just shipping features but defining the problem, choosing the metric, and iterating post-launch. The 3.6/5 aggregate employee rating suggests the bar is real but not performative; people who clear it tend to stay.

A practical pattern emerges from the guides: prepare a 90-second career narrative that ends by tying your experience to SentiLink's fraud mission, a 60-second answer on why this domain, and a 60-second answer on why this company. Have three to five STAR stories ready that show you led a measurable change: reduced false positives by X%, rebuilt a pipeline to cut latency by Y%, closed a complex enterprise deal in a regulated vertical. In the final round, candidates who proactively message the hiring manager or a future peer before the culture check — a short, specific note referencing something learned in earlier rounds — consistently report better outcomes. The correlation is strong enough that skipping it looks like an unforced error.

Where the work happens: two hubs, a distributed model

SentiLink's physical footprint reflects a company that grew up during the remote-work shift and never fully anchored to a single headquarters. The company's own contact page lists 33 New Montgomery Street in San Francisco's Financial District as its address, with a 415 area code phone number. Yet BuiltIn's office directory identifies 807 Brazos Street in downtown Austin as the headquarters and counts nine total locations. Craft.co shows only a single office at 171 2nd Street, also in San Francisco. The discrepancy isn't unusual for a distributed company; it suggests the legal domicile and the operational center may differ, or that data sources capture different snapshots of a changing footprint.

What's consistent across sources: San Francisco and Austin are the two primary U.S. hubs. The San Francisco presence appears at two addresses — 33 New Montgomery (the contact page) versus 171 2nd Street (Craft.co) — both in the SoMa/Financial District corridor where fintech and identity-verification companies cluster. Austin's 807 Brazos Street sits in the central business district, walking distance from the Capitol and the growing tech corridor along Congress Avenue. Between them, these two cities account for the largest share of SentiLink's 170-person headcount, including the 71 product and engineering employees BuiltIn tracks.

Beyond the two main hubs, the office map fans out into a mix of permanent sites and flexible arrangements. BuiltIn lists San Ramon and Seattle as U.S. locations: San Ramon in the East Bay's Bishop Ranch office park, Seattle likely serving the Pacific Northwest talent pool. Two California coworking spaces appear explicitly: a "coworking space in the Encino area" of Los Angeles' San Fernando Valley, and a "nice coworking space in SoHo" (the latter presumably in New York City). These read less like strategic expansions and more like satellite footholds for sales or partnership teams near key financial-industry centers.

Internationally, SentiLink maintains offices in Bengaluru and Gurugram, India, two distinct cities in the National Capital Region and Karnataka respectively. Bengaluru's concentration of ML and data-engineering talent makes it a natural fit for a company whose core product leans heavily on synthetic-identity detection models. Gurugram, adjacent to Delhi, hosts a deep bench of backend and platform engineers. Together they signal that SentiLink's technical center of gravity isn't purely American; the India offices likely carry meaningful product and engineering load, not just support functions.

The hybrid policy documented in public HR resources describes eligibility subject to department-head approval and role requirements. But SentiLink's own BuiltIn profile states it more plainly: "SentiLink supports a variety of ways to work, ranging from fully remote to in-office." That phrasing matches the job-board reality. The first-party Zero G Talent board shows roles tagged "United States" without a city requirement for positions like Head of Strategic Sales, Account Executive SLED East/West, and Head of Crypto GTM; all senior, high-compensation roles ($275k–$550k bands) that would traditionally demand physical presence. The implication: physical offices exist as collaboration anchors, not attendance mandates.

For a candidate evaluating where they'd actually sit, the picture is this: two substantial U.S. hubs with permanent leases in San Francisco and Austin, two established engineering centers in India, a handful of smaller U.S. outposts (some coworking, some leased), and a default-to-remote posture for roles that don't require face time with financial-institution partners. The offices are real, but they're not the story — the distributed model is.

Who thrives: the profile that stays and advances

The people who stay and advance at SentiLink share a recognizable profile: they want their work to map to a concrete problem, they operate with low ego across disciplines, and they treat autonomy as a responsibility rather than a perk. The company's stated mission — stopping identity fraud at the application stage for hundreds of financial institutions — attracts engineers, data scientists, and fraud analysts who measure success in prevented losses rather than shipped features alone. Built In's employer profile notes that many employees describe pride in tackling problems with measurable impact, and the weekly company-wide fraud case reviews (the same Thursday ritual Harris instituted at the start) reinforce that orientation by forcing every function to speak the same language: here is the attack, here is the signal we caught, here is what we missed.

Collaboration at SentiLink is structural, not aspirational. The profile describes colleagues as smart, reliable, and respectful, working closely across fraud intelligence, data science, product, engineering, go-to-market, and customer teams. That cross-functional rhythm shows up in the OKR operational model and team-based strategic planning, both documented in the employer profile, and in an open-door policy that leadership actually uses. Glassdoor reviewers rate culture and values at 3.6 out of 5, but the more telling signal is the 59 percent who would recommend the company to a friend, a figure that survives the 2.8 work-life-balance score, suggesting the trade-off is understood and accepted by the people who fit.

Autonomy appears repeatedly in the research: employee feedback shapes policies and strategy; managers run consistent feedback loops and public shoutouts; mistakes are framed as learning opportunities. The promotion framework makes this explicit: documented career ladders, internal-first job posting, lateral mobility encouragement, customized development tracks, and scheduled review cycles. In practice, that means a data scientist can rotate into a fraud intelligence role, or an engineer can move toward product, without leaving the company. The 71-person product and tech team is small enough that those moves are visible and supported.

Flexibility is real but bounded. The employer profile lists a full-time remote-friendly model, async-friendly policies, flexible schedules, and home-office stipends; yet it also emphasizes a strong in-person office culture with all-hands and revenue kickoffs held face-to-face. The nine offices, anchored by Austin HQ, exist for the people who want them; no one is required to commute, but the option is maintained and resourced. Candidates who thrive tend to self-select into the rhythm that matches their life stage: remote contributors who over-communicate in writing, hybrid workers who anchor their weeks around the case reviews, or office regulars who use the open floor plan for the spontaneous debugging sessions the culture encourages.

Technical depth is non-negotiable. The product pairs machine learning with human case reviews, which means engineers and data scientists must be comfortable shipping models that a fraud analyst will interrogate the next morning. That feedback loop — model output, human judgment, model retraining — selects for people who document their assumptions, welcome scrutiny, and iterate fast. The blog at resources.sentilink.com, where the team publishes data-driven fraud observations, doubles as a cultural artifact: writing clearly about messy adversarial problems is part of the job.

In short, SentiLink rewards mission alignment backed by craft, low-ego collaboration backed by structure, and autonomy backed by accountability. The 2.8 work-life-balance rating is the honest cost of that combination; the 3.6 culture score and 59 percent referral rate are the honest return.

The job board today shows seven sales roles. What doesn't change is the Thursday case review, the three million identities, and the median engineer who stays because the work maps to a problem that matters.

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