Who Gets Hired
Braintrust Data has paid more than $403.9 million to over 2 million vetted professionals across 100-plus countries while building the evaluation infrastructure that frontier AI labs and Fortune 2000 enterprises now run in production. This guide maps who Braintrust recruits, what they're paid, how the multi-stage interview loop evaluates candidates, where work happens across its San Francisco and Singapore hubs, and which traits let people thrive inside a company that operates both a global talent marketplace and the AI infrastructure powering it.
The roles on Zero G Talent's board cluster around five functions. Developer Support Engineer appears in both San Francisco and Singapore, signaling a global support layer. Software Engineer, Developer Experience sits beside it, showing heavy investment in SDK and API friction. A Data Engineer posting in San Francisco points to internal pipelines and customer-facing analytics. On the go-to-market side, a Strategic Account Executive for the West Region indicates enterprise sales motion, while a Head of Finance role rounds out the leadership bench.
Braintrust describes its sweet spot as "AI engineers, platform teams, and product organizations that need rigorous evaluation and production observability for agents and LLM applications rather than ad-hoc prompt testing." That framing matches the hiring pattern: these are not generic full-stack roles. They specialize around developer experience, data infrastructure, and the peculiar demands of shipping evaluation tooling that must run reliably in customers' production environments. The Singapore support posting further implies a follow-the-sun model, which only makes sense if the customer base is already global and the product in active daily use.
Candidates who have built developer tools, run evaluation pipelines, or sold technical infrastructure to platform teams will recognize the profile. Those who haven't need equivalent depth in adjacent domains.
The Pay Picture
Braintrust's salaried compensation clusters around a board-wide band of $115k–$175k, with a $155k median across the two roles disclosing ranges on Zero G Talent.
| Role | Location | Range |
|---|---|---|
| Developer Support Engineer | San Francisco | $157k–$180k |
| Developer Support Engineer | Singapore | 140k–166k SGD (~$103k–$122k) |
The San Francisco figure sits at the upper end of the company's band. The Singapore range reflects local market calibration, not a flat global grid. Both map to the same function: helping customers integrate and troubleshoot Braintrust's evaluation and observability tooling.
Other open salaried roles (Data Engineer, Strategic Account Executive (West Region), Software Engineer (Developer Experience), and Head of Finance, all in San Francisco) publish no ranges. The $115k–$175k band is the only public anchor until the company or candidates disclose more.
Equity does not appear in board listings. Candidates should ask recruiters for the current 409A price and share count; those details are negotiated per offer.
Benefits follow the well-funded startup template. The Singapore role adds CPF contributions and relocation support where applicable.
Targeting a Developer Support Engineer seat gives you an explicit range at the band's upper end. For engineering, product, sales, or finance roles, that band is the only public guardrail: come prepared with competing offers or market data to anchor the conversation.
Inside the Interview Loop
Braintrust publishes no interview playbook. Its careers site keeps stage-by-stage mechanics private. The only public first-person account tied to the brand is a 2021 YouTube walkthrough from a Braintrust marketplace participant describing how the platform evaluates freelancers. That video is not an employee hiring guide, but it reveals evaluation vocabulary the organization uses when assessing technical talent.
The first touch is written, not live
The 2021 video states: "the interview actually starts when your first message is sent… you want to always try to respond within 24 hours." The salaried postings, each listing a San Francisco or Singapore location and a narrow salary band, signal that the company expects applicants to self-qualify on geography and compensation before a human reviews the file.
Screening for signal
The marketplace video breaks down what the screener scores: preparation (company and role research), relevant experience framed as "mental check marks" rather than chronological narrative, and the ability to hook the listener in the first 60–90 seconds. It also flags appearance, audio clarity, and deliberate pacing: "slow down… enunciate even if it feels like you're talking slightly louder than you normally do." These are not formal rubric items for Braintrust's internal roles, but they match the profile the board postings imply: engineers who communicate across time zones, support developers in real time, and own a customer-facing surface without hand-holding.
Technical depth tested in context
The Developer Support Engineer and Software Engineer (Developer Experience) roles carry tight bands — $157k–$180k in San Francisco, 140k–166k SGD in Singapore — implying a calibrated leveling system. The Data Engineer and Strategic Account Executive listings suggest parallel tracks. No public leak describes the exact prompt library, but the company's own product (an evaluation platform for AI systems) makes it likely that at least one round asks candidates to critique or design an eval harness.
Disqualifiers the marketplace surfaces
The 2021 video notes a pattern among freelancers who win repeat engagements: "their commonality here is I want to be their friend… you talk to them on camera it's like oh tell me more… giving that vibe is so cool." The same source warns against rambling chronological answers, rushing without a pause to think, and treating the conversation as Q&A instead of dialogue.
What the board postings imply about cadence
Five distinct salaried roles listed simultaneously in San Francisco, plus a Singapore counterpart, indicate a hiring engine running multiple pipelines in parallel. The $155k median across two data points suggests offers cluster near the middle of the posted range. Candidates who advance typically show a portfolio artifact — a shipped SDK, a public eval framework, a postmortem they wrote — that the panel can dissect together rather than a rehearsed pitch.
Gaps in the public record
Braintrust has released no official interview guide, scorecard, or timeline. The 2021 marketplace video is the only that account tied to the brand, and it predates the current salaried hiring push by years. Treat the above as informed inference from the roles on this board and the evaluation language the company's own platform encourages. If you apply, ask the recruiter for the current stage count and evaluation criteria; they hold the authoritative version.
Where the Work Happens
Braintrust's postings reveal two primary hiring locations: San Francisco and Singapore. That role appears in both: San Francisco at $157k–$180k, Singapore at 140k–166k SGD. Additional San Francisco the previously listed roles, Head of Finance. The concentration suggests engineering and go-to-market functions anchor at Bay Area headquarters, with a growing Singapore presence.
Beyond those cities, the research contains no detail on physical offices, labs, test environments, or collaboration spaces. No floor plans, equipment lists, hardware testbeds, or workspace descriptions appear in company materials or employee accounts.
This absence matters for candidates. A data-intensive product organization typically relies on internal tooling: observability stacks, staging environments mirroring production, GPU clusters for model evaluation, or hardware-in-the-loop rigs if edge deployment is involved. Whether Braintrust maintains dedicated red-team labs, chaos-engineering environments, or simply leans on cloud-based ephemeral infrastructure isn't documented. The Singapore posting hints at a regional hub, but whether it houses engineering talent, sales support, or both isn't specified.
What can be inferred comes from the role mix. Those roles imply a product shipping APIs, SDKs, or tooling used by external developers, roles that usually sit close to the engineering teams building those surfaces. Strategic Account Executive (West Region) suggests a sales motion benefiting from proximity to product. Head of Finance in San Francisco signals the corporate backbone remains there. But inference is not evidence.
Candidates should ask directly: what does the San Francisco office provide that a remote setup doesn't? Is there a hardware lab, a device farm, or a dedicated staging cluster? Does Singapore have engineering seats or only go-to-market? The board data gives you the cities; the offer conversation has to yield the rest.
Who Thrives Here
Braintrust operates a two-sided talent marketplace that has paid out that amount to them globally, while simultaneously building the AI infrastructure (Brainstore, the Loop agent, MCP server, and the AIR interviewing product) that powers its own vetting and the workflows of those organizations.
Technical depth paired with product orientation. The engineering team ships a database (Brainstore) engineered for nested agent traces, a full-text search layer that outperforms generic competitors on latency, and an MCP server that lets coding agents query logs and update prompts directly from an IDE. The same crew runs eval loops that turn production traces into regression datasets with one click. The board's salaried roles (Developer Support Engineer, Software Engineer (Developer Experience), Data Engineer) all sit at the boundary where deep systems knowledge meets the ability to explain and iterate on behalf of customers.
Iterative, eval-driven mindset. Braintrust's public roadmap centers on the Loop agent: describe the optimization target, and the system generates better prompts, scorers, and datasets automatically.
Rigor without bureaucracy. The marketplace's founding insight: staffing agencies added 40–60 percent markup for almost no vetting value. Braintrust replaced that with "rigorous, skills-based vetting — not resume parsing" and later productized the tooling as AIR, which passed an independent third-party bias audit with zero adverse findings.
Customer empathy expressed through system design. G2 reviews repeatedly call out the "natural and empathetic" voice of the AI interviewer and note that candidates "feel genuinely at ease." The same empathy shows up in the enterprise product: Nexus connects 100-plus integrations so that workflows that once took weeks now take seconds.
Autonomy calibrated to distributed reality. The company scaled during the 2020 remote-work surge and now lists roles in San Francisco and Singapore with no relocation requirement. The marketplace itself is a distributed network of 2 million-plus vetted professionals.
Comfort with the AI-human complementarity thesis. Leadership states plainly: "AI infrastructure and human expertise are not competing forces — they're complementary ones." The fastest-growing segment of the marketplace is AI training infrastructure, where domain experts in healthcare, engineering, legal, and the humanities conduct model evaluation and preference ranking at scale.
The through-line is ownership of the full loop: write the code, ship the eval, watch the production traces, surface the pattern, update the scorer, repeat. Braintrust's tooling makes that loop fast; the people who thrive already run it in their heads before they open IDE, whether they're sitting in a San Francisco office, a Singapore hub, or the next city the marketplace reaches.
Working in frontier tech? Zero G Talent tracks the openings: see every open Braintrust Data role, browse frontier tech jobs, the companies hiring, and the people building the field.