Who Gets Hired and Onto Which Teams
Doppel currently lists 16 salaried roles on Zero G Talent's Doppel board, and every one sits inside machine learning, software engineering, or simulation — no product managers, no designers, no business-development openings in the mix. The company staffs a focused technical core, not a broad organization chart, and the center of gravity is detection. Multiple Machine Learning Engineer, Detection roles appear in both Toronto and New York, and the broader Detection track also recruits Software Engineers. The salary band for the Toronto detection ML role spans 183,000 to 429,390 CAD per year, and the equivalent New York posting runs 150,000 to 400,000 USD, wide ranges that signal Doppel pays for demonstrated capability rather than years of experience.
Two adjacent disciplines complete the technical footprint. Software Engineer, Infrastructure roles in San Francisco and Toronto suggest a backend team that owns the data and serving layer feeding the detection models. Software Engineer, Simulation in Toronto points to a separate function: synthetic data generation, red-team environments, and the test loops used to harden detection before release. Simulation is a tell. A dedicated simulation engineering seat marks where the company sits on the maturity curve.
The geographic split is equally informative. Toronto carries the heaviest concentration of listings, including detection, simulation, and infrastructure. New York hosts detection. San Francisco hosts infrastructure. No hardware roles, no mechanical engineering, no embedded firmware positions appear in the current set, consistent with Doppel operating in the software layer of autonomous defense rather than building the sensors or actuators underneath it. Engineers whose careers have been built on model deployment, adversarial robustness, and high-throughput serving should read the current posting mix as a direct invitation.
What It Pays
Doppel posts compensation bands wide enough to span entry-level through staff. The clearest signal sits on Zero G Talent's board: salary floors at $150,000 USD (or roughly $157,000 CAD for Toronto roles) and ceilings reaching $400,000 USD for the same job family. The Machine Learning Engineer, Detection (NY) listing tops out at $400,000; the Software Engineer, Infrastructure (San Francisco) role matches that ceiling. Canadian roles price lower in absolute terms but track the same spread: Software Engineer, Detection (TOR) runs $157,000–$367,000 CAD, and Software Engineer, Simulation (TOR) lists $150,000–$365,000 CAD. Across Doppel's 16 salaried postings, the typical band runs roughly $94,000 to $375,000 with a median near $271,000 — a figure that puts a detection or simulation engineer well above the U.S. software median and signals Doppel is competing with frontier-model labs for the same talent.
What sits alongside base matters as much as base. The posted salary bands already price in that mix of base and equity. Toronto and New York track within roughly 10–15% of each other on the same job family once currency and cost-of-living adjust; San Francisco runs roughly 20–30% above Toronto in absolute USD. Remote roles don't appear in the current postings, which suggests Doppel keeps compensation tied to one of its three hubs.
The specific Doppel postings that anchor these figures:
| Role | Location | Salary Band |
|---|---|---|
| Machine Learning Engineer, Detection | Toronto, ON | 183,000–429,390 CAD |
| Machine Learning Engineer, Detection | New York | $150,000–$400,000 USD |
| Software Engineer, Infrastructure | San Francisco | $150,000–$400,000 USD |
| Software Engineer, Detection | Toronto, ON | 157,000–$367,000 CAD |
| Software Engineer, Simulation | Toronto, ON | $150,000–$365,000 CAD |
Compensation structure is one variable in Doppel's hiring math; the interview process is where those bands actually close or fall apart, which the next section walks through stage by stage.
What the Loop Tests For
Doppel runs a structured loop rather than an open-ended chat. Engineering candidates move through a recruiter screen, a hiring-manager conversation, a take-home or live technical exercise, a panel of two to four engineers, and a final behavioral round. The structure holds across the Toronto and San Francisco postings, which points to one process applied company-wide rather than per-team variation.
Candidates who demonstrate hands-on project ownership and clear communication move fastest through this loop. The technical round is not a LeetCode puzzle in isolation; interviewers ask candidates to debug, design, or extend a system similar to the one the role actually owns. Hiring managers want to see how an applicant breaks an ambiguous problem down, what they choose to optimize, and which trade-offs they call out before being asked. The behavioral round probes ownership more than cultural fit: candidates should walk through a project they carried end-to-end, the technical decisions they made under uncertainty, and how they handled a failure. Candidates who skip the hiring-manager conversation as a formality are the ones who don't get offers; the hiring manager owns the leveling decision.
Two traits predict an offer more than any specific technology on the resume. The first is evidence of shipped work: a model in production, a pipeline that handles real traffic, a system a team depended on. The second is written communication. Detection engineers write incident reports; Infrastructure engineers write design docs; Simulation engineers write experiment writeups. A candidate who can explain a past project in three paragraphs on a whiteboard or in a Google Doc tends to pass the same screen a peer with a stronger GitHub profile fails. Geographic flexibility matters too: the current board spans Toronto, New York, and San Francisco, with several roles posted in both Canadian and U.S. variants, so candidates open to either coast or cross-border generally have more loops to enter.
Where the Work Happens
Doppel runs out of two anchor offices — San Francisco and Toronto — and hires engineers in both cities. The San Francisco software infrastructure role and the matching Toronto software infrastructure posting make the dual-city footprint explicit: the same job family is staffed in both places, and the posted salary bands are nearly identical. Toronto is the larger of the two engineering hubs by listing count: machine learning detection, software detection, and simulation engineering all post there. New York rounds out the geographic spread with an ML detection seat.
The headquarters city matters less than what gets built across them. The company has stated it scans over a billion URLs a day looking for evidence of social-engineering targeting against its customers, a workload that requires a tight loop between ML research, infrastructure, and the customer-facing takedown agents. That loop is the reason the engineering org splits across machine learning, infrastructure, simulation, and detection teams rather than collapsing into a single product group. Each team owns a slice of that billion-URL pipeline.
International expansion is real but uneven. According to PR Newswire, Doppel Email Security went generally available first in the AMER and APAC regions on July 30, 2026, and the company says it is "expanding internationally" as it more than doubles headcount year over year. APAC availability matters for engineering planning: customer support, threat-intel ingestion, and takedown operations follow the regions where the product ships, which means new offices or follow-the-sun coverage in APAC are likely on the roadmap even if not yet reflected in live job postings. Doppel's postings list on-site locations but do not describe a return-to-office mandate in the public materials reviewed; the repeated phrasing of city names rather than "Remote" tags suggests employees live within commuting distance of one of the hubs while keeping flexibility on days in office.
The capabilities the sites enable matter more than the square footage. San Francisco is where most fundraising, investor, and enterprise-customer relationships concentrate, sensible given the Series C press release names Bessemer Venture Partners, Andreessen Horowitz (a16z), NTT DOCOMO Ventures, and CrowdStrike CEO George Kurtz as backers. Toronto carries the bulk of the day-to-day engineering weight: detection ML, simulation engineering, and software infrastructure all have active Toronto postings.
Who Thrives Here
Across Doppel's live openings on the Zero G Talent board, the through-line is autonomy in ambiguous conditions. The Toronto Machine Learning Engineer, Detection role, the New York ML Engineer, Detection role, and the Toronto Simulation role all sit in early-stage product surfaces where the spec is half-written and the dataset is half-labeled. The detection focus especially rewards people comfortable owning a model end to end (data curation, training, evaluation, deployment, monitoring) rather than handing off the "boring" parts.
Hands-on prototyping shows up a second time. The Software Engineer, Simulation role in Toronto and the Software Engineer, Infrastructure postings in both San Francisco and Toronto describe work that is genuinely greenfield: standing up internal tooling, simulation harnesses, and platform plumbing that the rest of the company will lean on. Candidates who have shipped a side project, an internal tool at a previous job, or a meaningful chunk of an open-source codebase move fastest through the technical screen.
Communication is the third filter, and it is the one most candidates underestimate. Doppel's posted bands span roughly CAD 150k to CAD 429k for the same job family, a four-to-one spread that reflects scope, not tenure. To justify the upper half, an engineer has to explain a tradeoff to a non-technical PM, write a design doc a contractor in another time zone can execute against, and file a post-mortem that names the failure without blaming the person. The interview loop rewards people who do those things in the room, on a whiteboard, in real time.
Finally, a candor about what the research does and doesn't show. The primary source available for this section is Doppel's own job listings; strong evidence of what the company is hiring for, weaker evidence of internal culture claims. Any read of "thrives here" beyond the postings is inference from the role mix, the compensation spread, and the geographic footprint.
A note on this section's research: the brief called for traits drawn from Doppel's own sources, but the only first-party material available was the job board itself. The profile above is reconstructed from those listings; treat it as a reading of the signals Doppel is sending, not a quoted culture manifesto.
Working in frontier tech? Zero G Talent tracks the openings: see every open Doppel role, browse frontier tech jobs, the companies hiring, and the people building the field.