Eightfold AI

Eightfold AI is a talent intelligence platform founded in 2016 by Ashutosh Garg and Varun Kacholia, two engineers who came out of Google and Facebook respectively. Garg had led Search and Personalization at Google and IBM Research before co-founding BloomReach; Kacholia was an early Facebook engineer. The company raised a $220 million Series E in June 2021 led by SoftBank Vision Fund 2, which valued Eightfold at $2.1 billion and effectively doubled its valuation in six months. The platform's mission is to match people to work based on skills and potential rather than credentials, using a deep learning model trained on over one billion career profiles to infer transferable capabilities that traditional ATS keyword filters miss.

The platform covers four modules: Talent Acquisition for AI-driven sourcing and screening, Talent Management for internal mobility and succession planning, Workforce Exchange for employee-facing career intelligence, and Resource Management for project-based staff matching. In 2025, Eightfold added an agentic AI Interviewer that screens candidates around the clock and conducts preliminary structured interviews without recruiter involvement, alongside an AI Interview Companion that surfaces suggested questions and real-time response analysis during live interviews. The platform integrates with Workday, SAP SuccessFactors, Greenhouse, and Lever, and holds FedRAMP Moderate authorization and ISO/IEC 42001:2023 certification across all three levels. Customers include Bayer, STMicroelectronics, and large defense and financial services organizations.

What Eightfold AI actually is in May 2026

Eightfold sits at the intersection of applicant tracking, talent marketplace, and workforce intelligence. It is not an ATS replacement: it overlays existing HR systems as a skills and matching intelligence layer, ingesting data from Workday, SAP, Greenhouse, or any HCM the organization already runs. The core product is a proprietary skills graph built by analyzing job histories, role transitions, and career trajectories across more than one billion profiles. That graph allows the platform to infer that a marketer with content strategy experience maps to a sales role requiring lead nurturing, or that a departing engineer's skills map to an internal product role that would otherwise have gone to an external hire.

The four product modules operate on this shared skills foundation. Talent Acquisition uses it to rank candidates against job descriptions rewritten with skills criteria rather than years-of-experience proxies. Talent Management uses it to surface internal candidates for open roles before HR posts them externally, and to build development pathways that connect current employee skills to future role requirements. Workforce Exchange gives individual employees a self-serve view of their own skills profile and what roles inside the company they are positioned to pursue. Resource Management targets project-based organizations matching people to assignments based on skills availability rather than org chart hierarchy.

The agentic AI Interviewer, launched in 2025, automates top-of-funnel screening by running asynchronous structured interviews at any hour, dynamically adjusting follow-up questions based on candidate responses, and passing a ranked shortlist to human recruiters. This reduces time-to-interview for high-volume roles without adding headcount to the recruiting team.

Where Eightfold sits versus HireVue and Workday Skills Cloud

HireVue and Eightfold solve adjacent problems but operate at different stages of the talent funnel. HireVue concentrates on the interview and assessment layer: it evaluates candidates who have already been surfaced by running video interview responses and game-based assessments through industrial-organizational psychology models trained on more than 30 million data points. HireVue scores demonstrated competencies from structured interview answers, not inferred potential from career histories. Its compliance infrastructure (SOC 2, ISO 27001, FedRAMP) and audit trail capabilities make it the choice for legal teams worried about defensibility. Eightfold, by contrast, builds its advantage earlier in the funnel, sourcing and ranking candidates before any interview occurs, and extends further post-hire into mobility and succession. The two are often deployed together rather than competitively: Eightfold sources, HireVue assesses.

Workday Skills Cloud is a native capability inside the Workday HCM suite that builds a skills graph from data Workday already holds: job descriptions, performance reviews, feedback records, and compensation data. Its inference logic is narrower because it works from what is recorded in the enterprise system rather than from external career data. Workday knows that people with "software engineering" job titles tend to have Java associated because it appears repeatedly in their Workday data. Eightfold's external corpus of one billion profiles allows it to identify skills patterns and transferable adjacencies that never appear in a company's internal records, making it stronger for surfacing non-obvious internal talent and for hiring from outside. For organizations already on Workday that want modest skills intelligence without adding a vendor, Workday Skills Cloud is the path of least resistance. For organizations running a serious skills-based hiring transformation, Eightfold offers materially greater depth, at materially greater cost and implementation complexity. Users who pair both often use Eightfold's API to write enriched skills data back into Workday records.

A third point of comparison is Paradox AI, which focuses on conversational AI for high-volume hourly and frontline hiring. Paradox's Olivia chatbot automates screening conversations, interview scheduling, and onboarding paperwork, but does not operate a comparable skills graph or internal mobility layer. Where Eightfold targets white-collar workforce transformation at scale, Paradox targets speed and volume for shift-based roles. The two serve meaningfully different organizational profiles.

How AI actually works inside Eightfold

The platform's core intelligence comes from a proprietary deep learning model trained on career trajectory data rather than explicit skill tags. When a recruiter opens a requisition, Eightfold does not simply match keywords in resumes to keywords in job descriptions. It maps the requisition to a position in its skills graph and retrieves candidates whose career trajectories place them near that position, including candidates who have never listed a required skill explicitly but whose history suggests they have acquired it. This "inferred skills" mechanism is Eightfold's primary differentiator and its primary source of complaints: when the inference is wrong, a recruiter may receive a ranked shortlist of candidates who do not actually have the skill in question.

The diversity analytics layer applies profile masking during screening stages, withholding demographic signals to reduce bias at the point of ranking, while tracking diversity metrics across the funnel for HR reporting. The Talent Insights module runs workforce analytics that model current skills inventory against projected future requirements, flagging gaps that need to be addressed through hiring, reskilling, or acquisition before they become critical.

"Bayer is leveraging Eightfold's Talent Intelligence platform to drive a critical business transformation and enable talent flow at scale." - Manuel Smukalla, Head of Strategic Skills and Talent Flow, Bayer (Eightfold customer story, 2024)

Bayer expanded its internal talent marketplace from 16,000 to 100,000 employees globally in eight months using the platform, reallocating existing workforce skills to mission-critical projects rather than defaulting to external hires. STMicroelectronics reported saving over 160 hours of recruiting work in two months while scaling hiring efficiency.

"Eightfold increased our productivity per recruiter by over 50%." - James B., verified Capterra review, November 2019

The friction and legal risk users keep raising

Several recurring problems show up consistently across review platforms. ATS integration is the most common complaint: users report that connecting Eightfold to existing applicant tracking systems takes months and remains incomplete well into deployment. One reviewer noted their integration had not completed after starting the process in April. This is partly structural: Eightfold overlays existing systems rather than replacing them, which means bi-directional data syncing across different schemas is unavoidable.

Inferred skills create erroneous matches often enough that recruiters have to build manual calibration processes on top of the AI's output. The platform's confidence scoring for inferred skills is not always surfaced clearly, leaving recruiters to discover mismatch during phone screens rather than before. Soby A., reviewing on Capterra in August 2025, noted the platform is "not very versatile when it comes to customization and setup," and Mathieu L. called the UI and UX "really bad" as recently as February 2025.

The most significant concern to emerge in early 2026 is legal. On January 20, 2026, two job applicants filed Kistler et al. v. Eightfold AI Inc. in Contra Costa County Superior Court, California. The class action alleges that Eightfold harvested personal data on over one billion workers from social media, location data, and internet tracking, generated Match Scores on a 0-5 scale for job applicants, and automatically discarded low-scored candidates before any human reviewer saw their application, all without the disclosures the Fair Credit Reporting Act requires of consumer reporting agencies. Employers named in the suit include Microsoft, PayPal, Morgan Stanley, Starbucks, Chevron, and Bayer. Eightfold denies the allegations, stating the platform operates on data intentionally shared by candidates or provided by customers. The lawsuit has not been resolved as of May 2026, but it introduces compliance questions that any prospective buyer's legal team will need to assess before signing a contract.

Organizations evaluating Eightfold in the current environment should request clarity on how the platform handles FCRA compliance obligations and what disclosures are made to candidates whose profiles are scored, particularly in jurisdictions with strong consumer privacy frameworks. This issue is not unique to Eightfold but the litigation makes it concrete.

Who Eightfold is for

Eightfold is designed for large enterprises running deliberate skills-based hiring transformations. The sweet spot is organizations with 2,000 or more employees, multiple business units, high enough hiring volume to justify the integration investment, and a CHRO or People Analytics leader who has executive buy-in to restructure how roles are defined and filled. Industries that use it most are life sciences, defense contracting, financial services, and professional services, partly because these sectors have both the headcount and the regulatory compliance infrastructure that aligns with Eightfold's FedRAMP and ISO certifications.

Internal mobility is where Eightfold generates its clearest ROI signal. Organizations with large workforces experiencing high external hiring costs and employee retention problems find the platform's ability to surface internal candidates for open roles compelling, particularly when the alternative is paying search fees for external hires who leave within two years. The workforce planning layer adds value for CHROs modeling the skills impact of planned divestitures, acquisitions, or market pivots.

Skip Eightfold when your organization has fewer than roughly 2,000 employees, when you need a full ATS replacement rather than an overlay, or when your integration timeline cannot absorb a multi-month implementation. Startups and growth-stage companies will find the pricing, complexity, and minimum viable scale requirements all misaligned with their needs. Sourcing-focused teams expecting a drop-in replacement for LinkedIn Recruiter will also be disappointed: several users note the platform's candidate database still underperforms LinkedIn for niche or senior technical roles, and it functions best as a complement to existing sourcing tools rather than a replacement. For those use cases, tools like Clay for outbound recruiting research or lighter-weight candidate intelligence platforms will fit better.

For teams evaluating experience quality and user research tooling alongside talent infrastructure, UserTesting AI serves the complementary organizational intelligence picture.