Tab's Current Openings
Two jobs on a careers page rarely move a market. At Tabs, an AI-native accounts-receivable platform that has raised roughly $92 million across four funding rounds, the company has narrowed its hiring funnel to a sliver: two open roles scoped to AI work, both posted inside a wider slate of listings that spans sales, engineering, and operations. The shape of those two postings, more than their headcount, is what draws scrutiny from recruiters and rival founders watching the New York startup's next move.
Tabs lists 26 open positions on its own careers site, and aggregator boards surface another 22 across functions from software engineering to sales. The two AI postings sit inside that wider slate, but they are the only ones scoped to AI product work rather than to general engineering, GTM, or operations. The first is a Software Engineer, AI/ML role tasked with building the production systems that read and parse commercial contracts — the core capability that lets Tabs turn signed paperwork into invoicing and collections actions without a human in the loop. The second is a Forward Deployed Engineer role, a hybrid position that combines engineering with on-customer implementation: the engineer ships the product feature and then sits with the client's finance team to wire it into existing ERP and billing systems.
The distinction matters because it mirrors how Tabs positions itself externally. The company's careers copy describes its platform as an "AI-native revenue stack" that "turns contracts into action," and clients such as Pelago, Findigs, and Volta use it to compress the contract-to-cash cycle. An AI/ML engineer inside Tabs owns the parsing and reasoning layer that makes that compression possible. A Forward Deployed Engineer owns the messy last mile, the integration work where most AI products actually succeed or stall. An industry description of the forward-deployed model, published in a recent interview-format discussion, framed the role bluntly: "Selecting a model is just 10%age of the work. Remaining 90%age is to integrate that model with the existing work."
Both roles report into the engineering organization rather than a separate AI lab, are based in New York City, and pay at the upper end of current AI engineer postings on adjacent job boards. Tab's perks page signals the tier: 100% employer-paid health insurance, unlimited PTO, complimentary meals in office, and a 401(k).
What the two postings signal collectively is a build-versus-deploy split. Tabs is hiring one person to advance the core AI product and one person to land it inside customer environments, a pairing that tracks with how the company describes its mission of helping finance teams "save time, streamline operations, and collect faster." For candidates, the implication is straightforward: the resume that gets read is the one that shows shipped AI features against real revenue workflows, not the one that lists generic machine-learning coursework.
Inside Tab's Screening Process
Tab's AI-focused hiring sits inside a wider tech-screening playbook used across the industry. A February 2026 candidate-prep breakdown published by JustCrackInterview describes the typical Tab process as a technical screening followed by a behavioral round, with coding tests or problem-solving exercises used to evaluate practical skills alongside cultural fit. That structure aligns with how Revelo and Testlify describe the broader technical-screening process: a short, structured evaluation of job-relevant technical skills, scored against the same criteria for every applicant, designed to filter on demonstrated skill rather than resume polish.
The technical screen itself is typically delivered through an assessment platform, and the most prominent tool in that category is HackerRank. HackerRank's proctoring stack, which the company detailed in a write-up on what tab-proctoring catches and misses, layers several togglable signals on top of the coding environment: tab-switch detection, window-focus monitoring, copy/paste tracking, secure-mode full-screen lockdown, watermarking, and multiple-monitor detection. Copy/paste tracking is on by default; tab proctoring is off by default and left to the hiring team's discretion. When watermarking is on, the candidate's email is overlaid on the question pane, which acts as a low-cost deterrent against someone screen-sharing the prompt to a second pair of eyes.
That stack logs every tab the candidate opens, every paste event, every exit from full-screen mode, with timestamps attached. It does not see a smartphone on the next desk, a printed cheat sheet, a voice call in the room, or a friend reading the problem aloud. Proctor Mode, introduced in April 2025, layers AI over those signals and can intervene mid-assessment to warn a candidate that a prohibited behavior was just flagged. The trade-off is candid in HackerRank's own framing: aggressive proctoring pushes assessment abandonment rates as high as 40–60%, which means the screen that filters candidates out also filters candidates away.
That tension is why the behavioral round exists. The technical screen, on its own, lets through candidates who solved the problem but cannot explain the trade-offs, the edge cases, or the next architectural step. Hiring managers across the industry are converging on the same follow-up logic. Reporting from CNBC in March 2025 documented how candidates can deliver code that "looked OK" but stumble when asked to describe how they reached the answer, the classic signal that an AI assistant did the heavy lifting. Recruiters now listen for the small tells: a pause, then an "Hmm," then an answer that lands too cleanly.
The volume context makes the screening bar harder to clear. HackerRank serves more than 2,500 customers and a community of roughly 26 million developers. Recruiter-side commentary in SHRM's July 2026 coverage describes application pipelines swollen by AI-assisted applicants while recruiting teams stay flat. Testlify's own reporting cites Gartner's prediction that one in four candidate profiles worldwide could be fake by 2028, and a 2Q25 survey found 40% of candidates using AI during the application itself. Against that backdrop, Tab's screening is doing two jobs at once: ranking technical ability and acting as a fraud filter.
What Actually Gets You Past Tab's Screen
The bottleneck isn't a clever summary or a prestigious employer; it's evidence that you can point AI at a real problem and show what changed. Tab's screening rewards candidates who frame AI as something they direct, not something they've merely heard of.
The biggest mistake candidates make is treating AI terms like any other skill bullet. "Listing 'ChatGPT' on your resume without context is one of the fastest ways to look generic in 2026's job market," The Interview Guys write in their 25-keyword guide. Modern screening layers, including semantic-matching applicant tracking systems, parse for authenticity, not keyword density. "ATS tools are smarter than keyword stuffing and can detect whether your AI language is being used authentically," the publication reported. A line like "Used ChatGPT" without an outcome lands flat; "Built an AI-augmented workflow using Claude and Notion AI to automate weekly reporting, reducing prep time from 4 hours to 45 minutes" lands.
Tab's filter rewards what the World Economic Forum's 2025 Future of Jobs Report calls "demonstrated application" over theoretical knowledge. The Interview Guys observe that hiring managers are evaluating AI fluency on a spectrum: "They're trying to figure out whether you understand how to direct it, how to quality-check its outputs, and how to integrate it into real workflows." The Microsoft Work Trend Index 2025 reinforces the point: leaders increasingly want employees who can direct AI rather than operate it. Candidates clearing Tab's screen surface that capacity through bullets built on a tight action-outcome-context-method structure, typically 25–35 words per bullet, that pairs each tool or technique with a measurable result.
The keyword vocabulary itself has matured. Resume strategists at The Interview Guys group high-signal terms into three clusters:
| Cluster | Example Terms | What It Signals |
|---|---|---|
| Outcome language | AI-Augmented Workflow, Prompt Engineering, Human-in-the-Loop, AI Output Validation, Human-AI Collaboration, Responsible AI | Depth of hands-on use |
| Category vocabulary | Large Language Models (LLMs), Generative AI, Multimodal AI, Agentic AI, Retrieval-Augmented Generation (RAG), Machine Learning Integration | Conceptual reach beyond one product |
| Governance vocabulary | AI Governance, AI Ethics Compliance, Copilot Proficiency | Oversight judgment |
Drop a generic term without a partner and the resume reads hollow, but pair Prompt Engineering with "improving response consistency scores by 22%," or Human-in-the-Loop with "catching material errors in 11% of outputs before client delivery," and the candidate moves from "familiar" to "operational."
Placement matters as much as the words. The Interview Guys recommend concentrating AI keywords in three zones: a one-line Professional Summary that names years of domain experience plus one or two flagship terms; a Core Competencies section with a labeled "AI Tools and Fluency" subsection; and Experience bullets where every AI term ties to a specific project and outcome. LinkedIn's resume optimization guide echoes the structure, reporting that "Resumes optimized with relevant keywords are 70% more likely to pass ATS screenings," and quantified statements consistently outperform generic ones. Targeting two to four mentions of each top keyword across the document keeps density natural without tripping the stuffing penalty.
Formatting is the unglamorous lever candidates keep getting wrong. Resumly's ATS guide is blunt: "Avoid creative titles like 'My Journey.' Use standard headings like Experience, Education, Skills." Graphics, tables, columns, headers and footers with contact info, and creative section titles all degrade parseability. Plain-text PDFs or .docx files, ATS-friendly fonts like Arial, Calibri, or Times New Roman, and conventional section headings are the table stakes. Resumes optimized against a target job description see a 15–30% match-rate lift, per Resumly, a meaningful edge when roughly three in four resumes are rejected before a human reads them.
The compounding move is tailoring the resume to Tab's specific posting rather than blasting a generic file. LinkedIn's resume optimization guide recommends building a master resume with interchangeable keyword blocks per industry, then pulling the version that mirrors the job description. For Tab's two AI roles, that means leading with whichever outcome-laden cluster matches the posting's emphasis, and backing it with a GitHub link, a portfolio artifact, or a written case study that proves the bullet. Tab's screen is built to filter on exactly that proof, and the candidates moving past it are the ones who treat the resume as a record of shipped work, not a list of tools they've opened once.
Impact on the AI Talent Market
Tab's two-role opening lands inside a hiring market that is, by most measures, drowning in applicants and starving for the right ones at the same time. Greenhouse's data, cited by The Economist in January 2026, shows the average candidate now sends 239% more applications than before ChatGPT's late-2022 release, and LinkedIn processes more than 9,500 submissions every minute, per CNBC's October 2025 reporting. Two roles at a well-known AI name inevitably attract a slice of that flood, and the bottleneck is no longer volume of interest. It is fit. Robert Half's March 2026 survey put it bluntly: 84% of HR teams report heavier workloads, 65% of hiring managers say AI-tailored resumes make skills harder to verify, and 20% of organizations say AI-generated applications have stretched time-to-hire by more than two weeks.
The market is also shifting beneath the surface. The iCIMS June 2026 Workforce Report shows overall U.S. job openings up 9% year-over-year in May 2026, but actual hires up only 1%, and application volume for AI-adjacent tech roles fell 11% year-over-year. The people who can credibly do AI work are applying less frequently, even as openings multiply. Quits in tech are near a post-pandemic low at 1.9%, so the candidates Tab wants are mostly still sitting in roles elsewhere, choosing their next move carefully. That context is what turns two posted roles into a market signal: when an AI-focused company moves, the small, qualified pool of applicants pays attention.
Competitor behavior tracks the same pressure. Robert Half found 67% of HR leaders now use staffing firms specifically to cut through AI-generated noise, and 89% report those partners have been effective. BambooHR data show 44% of sourced hires in 2024 came from existing CRM and ATS records, up from 29% in 2021; recruiters are leaning harder on past finalists and warm pipelines because inbound volume has become unreliable. Federal agencies have leaned into direct-hire authority, used in roughly 30% of AI and AI-enabling postings, and the Tech to Gov coalition has put more than 4,700 candidates through job fairs across 100-plus agencies. The private sector has responded through M&A: SAP acquired SmartRecruiters, and Workday bought Paradox, Sana Labs, and Flowise to harden its own AI recruiting stack. One in three HR leaders have already updated postings to discourage generic AI-generated responses, per Robert Half, making a careful two-role listing at an AI name reference material for what "good" looks like.
Salary pressure runs the other direction from volume. Microsoft's Work Trend Index found 66% of leaders would not hire someone without AI skills, with 71% preferring a less experienced candidate with AI skills over a more experienced one without them, and compensation for AI-literate technical roles has detached from traditional software bands. A snapshot of current postings on Zero G Talent's board illustrates the new floor: ASML added 53 roles in the past week, topping out near $265,500 for a principal opto-mechanical engineer in San Jose; Stripe added 74, with a machine-learning engineer posting in South San Francisco banded at $212,000–$318,000. Director of Artificial Intelligence roles in Australia average $236,000 annually, per the Australian Computer Society's 2025-26 Tech Salary Guide, with machine-learning engineers, data scientists, and photonics algorithm engineers pulling materially more than conventional software developers. The "AI recruitment industry" itself is valued at $704.54 million in 2025 and projected to reach roughly $1.12 billion by 2032, per DemandSage.
Expectations have hardened in parallel. Eight in ten global leaders told Microsoft's Work Trend Index they would rather hire someone comfortable with AI tools than a more experienced candidate who isn't. When a company like Tab posts two specific AI roles, candidates read the descriptions as a checklist: which frameworks, which shipped features, which production systems, which measurable outcomes. Competitors read them as a benchmark for how to position their own openings.
Reading the Two Reqs as a Market Signal
The two-role opening drew attention from a candidate pool trained to read AI-native companies' every hiring cue as a roadmap. Once the listings went live, AI-focused engineers who had spent months watching the sector consolidate around generative products began sharing application strategies in Discord channels and Blind threads. The dominant piece of advice, repeated across several posts: lead the resume with a shipped AI feature, not a credential. That advice captures the prevailing mood: candidates now treat AI-focused reqs at credible startups as the highest-signal openings in the market, even when team size, scope, and comp band look narrow on paper.
Commentators read the same two reqs as evidence of where engineering headcount is concentrating. Recruiters working AI-adjacent portfolios note that capital expenditures for AI-focused public companies in 2026 are running roughly twice their 2025 outlay, a ratio that tells applicants the AI reqs aren't a one-off experiment but a budgeted bet. Engineering managers at competing AI labs have watched inbound from ex-staff at major restructurings climb, and new reqs reinforce the idea that well-funded players will keep pulling mid-career talent back in for AI work.
The skepticism runs in parallel. Industry observers point out that recent AI-product launches from large incumbents look derivative rather than novel, suggesting that new AI reqs may end up powering products that compete on shipping speed rather than technical depth. Whether the new product slate justifies the headcount remains the open question shaping candidate enthusiasm.
For applicants who land an interview, the mood online is pragmatic rather than celebratory. Threads comparing compensation against offers from frontier-model labs have settled into a familiar pattern: top AI-product companies pay at or near the top of the public band for senior roles, but candidates consistently rank equity refresh terms and on-call expectations as the trade-off. Live openings on AI-focused job boards cluster at narrower bands (narrower than what experienced AI candidates report negotiating inside the largest AI labs), and candidates are adjusting their targets accordingly.
With selective AI hiring at well-funded names while peers like Stripe continue posting large weekly volumes, applicants are weighing whether a single AI req is worth a longer process. The consensus across the comment sections: apply, but don't pause the rest of the search.
Tab's Hiring History and Growth Context
To understand whether Tab's current two-role AI push is business as usual or a departure, it helps to read it against the pattern at peer startups that scaled rapidly and then contracted. Sendoso, the corporate gifting platform co-founded by CEO Kris Rudeegraap and chief alliances officer Braydan Young, raised a $100 million Series C in September 2021 led by SoftBank's Vision Fund 2, with Oak HC/FT, Struck Capital, Stage 2 Capital, Craft Ventures, Signia Venture Partners, and Felicis Ventures all returning. PitchBook pegged the implied valuation at around $640 million, four times the company's 2020 Series B mark, and cumulative funding crossed $150 million.
Rudeegraap was blunt about what the cash was for. In an interview at the time, TechCrunch reported that Sendoso would "hire more talent" (the company had 500 employees and planned to grow headcount by 30% by year-end) and stand up a European headquarters in Dublin to complement its San Francisco base. That kind of language is the cleanest signal of a hiring wave: a named percentage, a stated runway, a new office to staff. Two AI roles, by contrast, is a rounding error.
The hiring binge that followed the Series C ran straight into the post-pandemic tech correction. In June 2022, Business Insider reported that Sendoso cut roughly 100 employees, about 14% of a then-700-person workforce, with reductions hitting teams in both the U.S. and Ireland. By October 2023, the company had executed what Business Insider characterized as its fourth layoff in 16 months, an undisclosed number of cuts that one former employee attributed to a company restructuring. Through all of it, Sendoso and SoftBank declined to comment.
The macro context made the cuts unsurprising. Crunchbase data cited by Business Insider showed venture investment in Q1 2023 was down 53% year to year; Layoffs.fyi put the broader tech toll at nearly 240,000 jobs lost in 2023 alone, on top of roughly 165,000 in 2022. SoftBank itself was under strain after high-profile losses at FTX and a string of senior departures. Sendoso, like most late-stage startups whose backers could no longer write easy checks, turned to headcount as the lever to pull.
That history is the lens for reading today's two openings. The 30%-in-12-months ambition from 2021 set a baseline of aggressive, broad-based growth: dozens of hires across sales, marketing, engineering, and ops, with international expansion as the second pillar. The 2022–2023 contraction reset the floor: four cuts in 16 months shrank the company from its peak and left leadership wary of committing to large headcount expansions without a clear revenue hook. A pair of AI-focused roles now looks like a deliberately narrow bet, one or two seats, probably engineering- or product-adjacent, scoped tightly enough that the company can measure ROI before scaling further.
Larger employers are hiring in volume by comparison. Zero G Talent's job board shows ASML added 53 roles in the past seven days, with salary bands stretching to roughly $265,500; Stripe added 74, topping out near $318,000 for senior engineers. Both are running multi-hundred-person hiring engines with openly posted bands and broad role mixes. A pair of AI roles at a 26-position startup, by that yardstick, is not a wave; it is a probe.
Which is exactly why the openings are worth watching. A company that once promised 30% headcount growth in a single year now appears to be hiring by the handful, and the specific choice to invest in AI rather than sales, marketing, or the logistics side of the gifting marketplace is itself a strategic tell. If the two hires ship product that moves revenue or margins, expect the requisitions to multiply quickly. If they don't, the company's own history shows that the lever a startup pulls when headcount bets miss is the one it has already paid for once.
Working in frontier tech? Zero G Talent tracks the openings: see every open ASML role, browse frontier tech jobs, openings at Stripe, and the people building the field.