AI and people intelligence terms to know.
This glossary explains the AI, data, and people intelligence concepts that shape modern people-centric work. It includes foundational definitions, workflow applications, and the signals and platform components that power Findem.
Last updated November 11, 2026
AI & Data Foundations
Artificial Intelligence (AI)
AI refers to systems that perform tasks requiring human-like reasoning, pattern recognition, or decision-making. In talent workflows, AI interprets data at scale to support sourcing, screening, analytics, and communication.
Machine learning (ML)
A subset of AI that uses algorithms to identify patterns and make predictions from data. ML powers tasks like candidate filtering, ranking, and classification.
Deep learning
A type of ML that uses neural networks to process unstructured or high-volume datasets — including resumes, profiles, and company information.
Generative AI (GenAI)
GenAI creates new content — outreach messages, summaries, job descriptions — by learning from large datasets of text and code.
Large language models (LLMs)
AI models trained to understand and generate human language. In talent workflows, they help interpret recruiter intent, answer questions, and power conversational interfaces.
Learn more about large language models
Person data sources
A person's professional footprint — roles, achievements, contributions, certifications, and digital profiles — used to generate structured talent data.
Company data sources
A company's digital footprint — funding events, team growth, markets served, leadership structure — that provides context for interpreting a person's experience.
3D People Graph
Findem's structured intelligence foundation, containing trillions of expert-labeled data points across individuals, companies, and time. It connects who someone is, where they worked, how their career evolved, and how they are connected to others — giving AI the context it needs to reason over people data, not just search it.
3D data (Person × Company × Time)
The data model underlying the 3D People Graph. It structures career history as three intersecting dimensions — person, company, and time — so the system understands what experience actually means in context, not just what titles a candidate held.
3D candidate profiles
Profiles generated from 3D data that provide an integrated view of a person's background, achievements, and career trajectory.
Attributes
Verifiable facts derived from 3D data — scope increases, technical depth, tenure at a specific company stage, industry specialization — used to assess fit with precision.
Findem Magics
Findem's library of 300,000+ proprietary, searchable attributes that translate intangible, hard-to-articulate career qualities — impact, leadership style, strategic thinking, retention risk — into concrete, data-backed search criteria. Magics close the gap between what a recruiter knows they want and what a keyword search can find.
AI Labeling Engine
The system that transforms raw person and company data into structured 3D data, identifies attributes, and produces Success Signals and Relationship Signals. It combines machine-scale processing with expert human review, so the intelligence behind every Findem recommendation is labeled, structured, and verifiable — not generated on the spot.
Learn more about the Data Labeling Engine
Success Signals
Verified, expert-labeled patterns that indicate which experiences and career markers predict success for a role, team, or environment. Success Signals are not black-box scores — they are structured, explainable patterns that help teams narrow the field with more confidence. Findem attaches roughly 75–100 Success Signals to a given candidate profile.
Learn more about Success Signals
Relationship Signals
Structured representations of how people are connected inside and outside an organization — through shared work history, alumni networks, referral paths, employee connections, and permissioned network relationships. Warm paths identified by Relationship Signals convert 2–8x more than cold outreach.
Learn more about Relationship Signals
Domain-specific AI for talent
AI models built on talent-specific data and expert-labeled signals, rather than general-purpose language models. Because these models are trained on role expectations, career patterns, and organizational context, they produce recommendations grounded in how hiring actually works — not just how language works.
People Intelligence Platform
A People Intelligence Platform turns fragmented people data into context that teams and AI can reason over and act on — across hiring, executive search, internal mobility, learning and development, and workforce planning. The intelligence is labeled, structured, and verifiable, not generated from general-purpose models.
One platform — three commercial forms. Apps (the customer owns the work), Agents (Findem does a bounded piece of the work and passes it back), and Services (Findem owns the result). The same expert-labeled intelligence and agent infrastructure underlies all three.
Learn more about the People Intelligence Platform
Insights powered by Findem
Analysis derived from 3D data, Success Signals, and Relationship Signals that supports workforce planning, hiring decisions, internal mobility, and talent strategy. These insights reflect structured, labeled data, not raw search output.
Explainable AI
Explainable AI (XAI) is AI that shows the evidence and reasons behind its outputs, in terms the person using it can understand. In hiring, that means a recruiter can see why a candidate was recommended or ranked, check that reasoning against the record, and decide whether to act on it.
Learn more about explainable AI
Human in the loop
Human in the loop (HITL) is a way of designing AI systems so that a person reviews, approves, corrects, or overrides the system's output before it takes effect. In hiring, that means a recruiter or hiring manager checks an AI recommendation, such as a ranked candidate or a drafted message, before it affects a candidate.
Learn more about human in the loop
AI recruiting agent
An AI recruiting agent is software that uses AI to plan and carry out multi-step recruiting work toward a goal, such as finding, contacting, and screening candidates for a role, with less step-by-step direction than an assistant. The recruiting team sets the criteria and permissions, reviews what the agent returns, and makes the hiring decisions.
Learn more about AI recruiting agents
NYC Local Law 144
NYC Local Law 144 is a New York City law that regulates automated employment decision tools (AEDTs) in hiring and promotion. Employers and employment agencies may not use a covered tool to screen candidates or employees unless it had an independent bias audit within the past year, a summary of results is public, and required notices were given.
Learn more about NYC Local Law 144
Data labeling
Data labeling is the process of adding meaningful tags, categories, or annotations to raw data, such as text, images, or records, so a machine learning model can learn from it or be tested against it. People, software, or both apply the labels, and each labeled example shows the model what the correct answer looks like.
Learn more about data labeling
Knowledge graph
A knowledge graph is a way of organizing data as a network of entities, such as people, companies, and places, connected by named relationships, such as "worked at" or "invested in." Because facts are stored as connections, software can follow paths between entities, combine data from many sources, and answer questions that span them.
Learn more about knowledge graphs
Entity resolution
Entity resolution is the process of finding records that refer to the same real-world person, company, or thing and linking or merging them, often when no shared unique ID exists. It is also called record linkage, deduplication, or data matching, and it lets teams count, search, and analyze each entity once.
Learn more about entity resolution
Skills intelligence
Skills intelligence is information about the skills people have and the skills work requires, gathered from many sources and analyzed so organizations can act on it. It shows which skills are in demand, where gaps and shortages exist, and how skills are changing, to guide hiring, training, internal mobility, and workforce planning.
Learn more about skills intelligence
EU AI Act
The EU AI Act is the European Union's law on artificial intelligence, Regulation (EU) 2024/1689. It sorts AI systems by risk, bans certain practices, and sets obligations for providers and deployers of high-risk systems. AI used to recruit or select people, or to make decisions about workers, is listed as high risk.
Learn more about the EU AI Act
Colorado AI Act
The Colorado AI Act is the common name for Colorado's law on automated decisions about consumers, including employment. First passed as SB 24-205 in 2024, it was repealed and reenacted by SB 26-189 in May 2026. The current law regulates automated decision-making technology and takes effect on January 1, 2027.
Learn more about the Colorado AI Act
Adverse impact
Adverse impact is a substantially different rate of selection in hiring, promotion, or another employment decision that works to the disadvantage of members of a race, sex, or ethnic group. Under the federal Uniform Guidelines, a group's selection rate below four-fifths, or 80%, of the highest group's rate is generally regarded as evidence of it.
Learn more about adverse impact
Data enrichment
Data enrichment is the process of adding missing, updated, or related details to records an organization already holds, usually by matching them against other sources. In talent work, it often means adding current roles, skills, and experience to candidate profiles in an applicant tracking system or CRM so the records stay useful.
Learn more about data enrichment
Data provenance
Data provenance is the record of where a piece of data came from and what happened to it along the way, including its original source, who or what created or changed it, and when. It lets people judge whether data can be trusted, trace an AI output back to its inputs, and correct errors at the source.
Learn more about data provenance
Disparate impact
Disparate impact is a theory of employment discrimination under Title VII of the Civil Rights Act. A practice that looks neutral, such as a test or education requirement, can be unlawful if it causes a disparate impact based on race, color, religion, sex, or national origin and the employer cannot show it is job related and consistent with business necessity.
Learn more about disparate impact
Illinois HB 3773
Illinois HB 3773 is the 2024 bill, now Public Act 103-0804, that amended the Illinois Human Rights Act to cover employers' use of artificial intelligence. Since January 1, 2026, it has been a civil rights violation for an Illinois employer to use AI that has a discriminatory effect or to fail to give notice of its use.
Talent Workflows & Use Cases
Talent decisions
An umbrella term for decisions across hiring, mobility, succession, retention, and development. Findem's platform supports these decisions with shared context and explainable signals.
Talent sourcing
The practice of finding qualified candidates and engaging them for open roles. AI supports sourcing by interpreting role expectations, ranking candidates, and identifying warm relationships.
Learn more about talent sourcing
Multichannel sourcing
A sourcing strategy that pulls talent from inbound applicants, referrals, rediscovery, alumni, and external search. Warm channels typically produce faster engagement and stronger pipelines.
Natural language sourcing
A sourcing capability that allows recruiters or hiring managers to begin a search using plain language. The AI interprets intent and translates it into search criteria.
Attribute search
Searching for talent based on verified attributes — such as company-stage experience, promotion velocity, scope growth, or industry depth — rather than keywords or job titles.
Copilot for sourcing
An assistive AI companion that helps interpret job descriptions, run searches, rank candidates, and accelerate outreach. Copilot works with human oversight and enhances efficiency.
Learn more about Copilot sourcing
Candidate rediscovery
Identifying qualified candidates already in your ATS. AI refreshes profiles with updated data and highlights past applicants who now fit active roles.
Learn more about candidate rediscovery
Candidate Relationship Management (CRM)
Nurturing and tracking qualified candidates for future roles. AI-driven CRMs personalize outreach, update data automatically, and support ongoing engagement.
Learn more about talent marketing and CRM
Talent ecosystem
A holistic strategy that combines sourcing, CRM, referrals, alumni, employer branding, and talent communities into one unified approach.
Talent analytics
Analytics that help teams understand funnel health, sourcing performance, recruiter activity, and outreach effectiveness. Findem unifies data across channels to improve insight.
Learn more about talent analytics
Talent insights
Data-driven analyses that support decisions related to hiring, planning, mobility, and development. They draw from 3D data, Signals, and observed behaviors.
Calibration (process)
Iterative alignment on what "great" looks like for a role. Calibration uses Success Signals, market data, and examples to ensure teams share expectations before sourcing begins.
Learn more about the Calibration Agent
Interview-ready
A state where candidates are pre-screened, qualified, and prepared to enter interviews. Interview-ready candidates include clear reasoning and documented signals.
Time to fill
Time to fill is the number of days it takes to fill an open position, usually counted from the day a requisition is approved to the day a candidate accepts the job offer. Talent teams average it across the roles filled in a period to see how long seats stay empty and where hiring slows down.
Intake meeting
An intake meeting is the conversation a recruiter and hiring manager hold before a search starts to agree on what a role needs. They confirm the duties, must-have and nice-to-have requirements, location, level, budget, assessment plan, and timeline, so sourcing, screening, and interviews all work from the same written brief.
Learn more about intake meetings
Skills-based hiring
Skills-based hiring is an approach that selects candidates on the job-related skills and competencies they can show, rather than on degrees, job titles, or where they learned. Employers define the skills a role needs through job analysis, then use assessments such as work samples and structured interviews to measure them.
Learn more about skills-based hiring
People analytics
People analytics is the practice of analyzing data about an organization's people to answer business questions and guide decisions about hiring, retention, development, and workforce planning. It is also called HR analytics or workforce analytics. Most of its data comes from internal systems such as the HRIS, payroll, performance reviews, and engagement surveys.
Learn more about people analytics
Talent marketplace
A talent marketplace, or internal talent marketplace, is a platform that matches an organization's current employees with internal opportunities such as open roles, short-term projects, gigs, stretch assignments, and mentoring. Matches are usually based on skills, experience, and career interests, so people can find their next opportunity inside the company before managers look outside.
Learn more about talent marketplaces
Workforce intelligence
Workforce intelligence is the combination of an organization's internal workforce data with external labor market data to guide decisions about hiring, locations, skills, and workforce planning. It answers questions internal reports cannot, such as where the needed skills exist, how competitors' workforces are changing, and whether the market can realistically supply the roles in a plan.
Learn more about workforce intelligence
Headcount planning
Headcount planning is the process of deciding how many people an organization needs, in which roles, teams, and locations, and when, within an approved budget. It turns business plans and financial forecasts into targets for new hires, backfills, and reductions, and it is usually shared between finance, HR, and the business leaders who own the work.
Learn more about headcount planning
Job requisition
A job requisition is a formal request to fill a position, created by a hiring manager and approved before recruiting begins. It records why the role is needed and details such as the title, team, location, and budget. Once approved, the requisition opens the role for recruiting, and many teams start counting time to fill from that point.
Learn more about job requisitions
Retained search
Retained search is a recruiting model in which a client hires a search firm exclusively and pays a fee that does not depend on a successful hire, usually in milestone installments. It is most common for senior and executive roles. In contingency search, by contrast, the recruiter is paid only when a referred candidate is hired.
Learn more about retained search
Skills gap analysis
A skills gap analysis compares the skills, knowledge, and competencies a workforce has today with the ones it needs now and in the future. It can cover a whole organization, a team, a role, or one person. The result shows where capability is short, so leaders can decide whether to train, redeploy, or hire people to close each gap.
Learn more about skills gap analysis
Skills taxonomy
A skills taxonomy is a structured, shared vocabulary of skills. It gives each skill a standard name and description, groups related skills into categories, and often links skills to occupations or roles. Organizations use one so that job descriptions, employee profiles, learning content, and workforce reports describe the same skill in the same way.
Learn more about skills taxonomies
Employee turnover rate
Employee turnover rate is the share of an organization's employees who leave during a period, usually a month or a year. It is the number of separations divided by average employment, times 100. Teams often split it into voluntary and involuntary turnover and track regrettable attrition, the loss of employees they wanted to keep.
Learn more about employee turnover rate
Applicant tracking system (ATS)
An applicant tracking system (ATS) is software that employers use to manage hiring from job posting to offer. It stores applications and candidate records, moves candidates through interview stages, and keeps feedback, approvals, and offer details in one place. It is usually the system of record for open roles and the people who applied.
Learn more about applicant tracking systems
Structured interview
A structured interview is a job interview in which every candidate for a role answers the same job-related questions in the same order and is rated on a common scale. Interviewers agree on what good answers look like before interviews begin and usually record ratings on an interview scorecard, so candidates can be compared on the same evidence.
Learn more about structured interviews
Silver medalist
A silver medalist is a finalist who was not hired for a role, usually the runner-up or another candidate who reached the last interview stage. Recruiters keep these candidates on record because they were already vetted for similar work, and they often contact them first when a comparable role opens.
Learn more about silver medalists
Employee referral program
An employee referral program is a formal process for current employees to recommend people they know for open roles. It sets how referrals are submitted and tracked, how referred candidates are reviewed, and whether employees receive a bonus or other recognition when a referral is hired. Referred candidates usually go through the same hiring process as other applicants.
Learn more about employee referral programs
Career pathing
Career pathing is the practice of mapping how employees can move between roles over time, including promotions, lateral moves, and moves into new job families. A career path lists the roles in that progression and the qualifications, competencies, and development activities needed at each stage, so employees and managers can plan the next step.
Learn more about career pathing
Span of control
Span of control is the number of employees who report directly to one manager. It is also called span of management or span of supervision. Together with the number of management layers, it shapes how flat or tall an organization is, how much attention each employee gets, and how much coordination each manager carries.
Learn more about span of control
Upskilling
Upskilling is short-term, targeted training that adds to, improves, or updates the skills employees already use, usually so they can do their current work better or take on more. Reskilling is training that gives people new skills for a different occupation or new professional activities. Both help organizations close skills gaps with the people they already employ.
Findem Platform &Agentic AI
Talent Solutions
Findem's direct enterprise applications — sourcing, CRM, screening, assessment, interview, verification, learning and development, and workforce planning — built on the 3D People Graph and delivered to talent teams as workflow tools. Each product is built on the same expert-labeled foundation so work stays connected across the talent lifecycle.
Embedded Solutions
Findem's co-innovation model. Job boards, HR platforms, and talent networks embed Findem's data and agent infrastructure to deliver AI-powered hiring outcomes their customers couldn't build on their own. Partners access the same expert-labeled intelligence and agent infrastructure that powers Findem's own products.
Findem Studio
Findem's AI-native work surface — people intelligence, built for AI. Studio turns that intelligence into finished work you can trust. Ask for a succession plan, a market map, a benchmark, or an intake, and it comes back done: built on labeled people data, following an expert methodology, and checked against the evidence before you see it. Three things make that possible: the right intelligence before it starts, the right method while it works, and the right checks before anyone acts.
Run a pre-built agent, build your own, or connect Studio's MCPs to whatever AI you're already using.
Learn more about Findem Studio
Model Context Protocol (MCP)
An open protocol that gives AI systems a connection to external data, tools, and capabilities. Findem exposes its people intelligence and agents via MCP so they can be called from Claude, other AI clients, and custom-built applications. MCP provides the connection — what Findem adds on top is the structured intelligence, expert methodology, and verification layer that turns raw access into finished work.
Findem Studio agents
Pre-built agents available in Findem Studio. Each returns a finished artifact — not a list or a score to still process.
Agents are grouped for Executive Search, Talent Acquisition, and HR Consulting. Run one directly in Studio, access it inside the Findem platform, or call it through MCP from the AI you already use.
Named expert methodology
Each Findem Studio agent runs a defined workflow authored or reviewed by a named practitioner — not a method the model invented on the spot. This is what makes Studio output defensible: a person with domain expertise reviewed the methodology before the agent ran it.
Intelligent Job Post
A job post that functions as an active agent. It reaches out to qualified talent, manages replies, conducts pre-screens, and builds an interview-ready pipeline automatically — attached to a single role.
Learn more about the Intelligent Job Post
Assistive AI
AI that accelerates work with human oversight. It supports planning, sourcing, analytics, and communication — but a person makes the final call at each step.
Learn more about the People Intelligence Platform
Agentic AI
Autonomous AI that plans, executes, and refines multi-step workflows to deliver outcomes. In talent acquisition, the primary output is interview-ready candidates.
Learn more about agentic AI in HR
Agentic workflow
A multi-step workflow in which an AI agent plans, executes, and refines tasks to deliver a defined outcome. In talent acquisition, the primary output is interview-ready candidates or a finished people-work artifact. Agentic workflows differ from assistive AI in that the agent handles sequencing and execution — a person reviews the output and makes the final call.
Learn more about agentic workflows
Calibration Agent
An agent that structures intake conversations, evaluates market data, and aligns hiring teams on Success Signals and examples before sourcing begins.
Learn more about the Calibration Agent
Application Boost Agent
An agent that increases completed, qualified applications through personalized outreach and a simplified application experience.
Learn more about the Application Boost Agent
Screening Agent
An agent that runs pre-screens using role-aware questions, analyzes responses, and returns ranked candidate summaries with documented reasoning.
Learn more about the Screening Agent
Scheduling Agent
An agent that finds availability, books interviews, manages reschedules, and keeps calendars and ATS statuses in sync.
Learn more about the Scheduling Agent
Assessment Agent
An agent that delivers role-aligned, proctored simulations, flags anomalies, and returns scores with standardized summaries.
Learn more about the Assessment Agent
ID Verify Agent
An agent that performs identity verification and credential checks to confirm candidate authenticity before interviews or hire.
Learn more about the ID Verify Agent
Veteran Sourcing Agent
An agent that surfaces qualified veteran talent from trusted communities and returns interview-ready candidates.
Learn more about Veteran Sourcing Agents
Fia (Findem Intelligent Assistant)
Findem's conversational AI layer and in-platform orchestration tool. Fia coordinates agents, manages workflow sequencing, and maintains consistent context across the hiring lifecycle — so each step informs the next. Embedded directly in product surfaces, Fia lets recruiters take action through plain-language requests while keeping humans in control of every consequential decision.
Fia never makes final calls: it shows evidence against criteria the recruiter set, and always requires explicit confirmation before executing bulk actions.
The Takeaway
Most AI in talent tools is general-purpose — trained on language, not on careers. It can generate text and move tasks along, but it doesn't understand what good looks like for a specific role, team, or stage of company. And MCPs have become commoditized: access to people data is now table stakes.
Findem is built differently. The 3D People Graph contains trillions of expert-labeled data points across people, companies, and time. Success Signals and Relationship Signals give the platform a structured view of how careers actually unfold and what patterns predict performance in a given context. And Findem Studio brings all of that together with a named expert's methodology and verification before anyone acts — so what comes back isn't raw material to still process. It's finished work you can defend.
That's what separates access from outcomes, and task automation from decisions you can act on with confidence.