GoodTime
Also known as: GoodTime Hire, Orchestra, Cori
Enterprise interview scheduling and hiring coordination platform whose agent, Cori, proactively screens candidates, books and repairs multi-day panels, and coordinates recruiters, interviewers and candidates across the existing ATS and calendar.
GoodTime is an enterprise platform for interview scheduling and hiring coordination, built for the complexity that defeats simpler tools, such as multi-day panels across time zones, interviewer load balancing and high-volume hiring events, and used by talent teams at Databricks, HubSpot, Lyft, Shopify, Spotify, Priceline, HelloFresh, Zalando, OLX and Glovo.
The platform runs on Cori, an AI agent GoodTime describes as a digital teammate, not a chatbot or copilot. Cori does not wait to be prompted. She screens and prioritizes candidates, schedules and reschedules interviews, coordinates interviewers and availability, converses with candidates in real time, sends updates and reminders, advances candidates through stages and surfaces bottlenecks. When an interviewer drops out she finds a replacement and tells everyone involved. She works over email, SMS, WhatsApp, Slack, Microsoft Teams, in-platform chat and voice, and operates inside the customer's existing ATS, calendar and video conferencing instead of replacing them.
Oversight is built around visibility and boundaries, not a universal approval step. Every action Cori takes is surfaced with the reasoning behind it, teams can step in, adjust or override at any point, and when a decision needs input or something falls outside the workflows the customer has defined, Cori escalates to a person. Personally identifiable information is stripped before AI processing, and a Bias Audit Report lets customers verify and continuously monitor that AI candidate matching performs fairly across demographic groups, with country-level reporting for regional analysis.
Security is unusually well evidenced for the category, pairing SOC 2 Type II, SOC 3, ISO 42001, HIPAA/HITECH Type 1, GDPR and CCPA coverage with SAML single sign-on across major identity providers and a four-level user permission model administered by the customer. Data is hosted on Heroku and AWS, with an EU-based hosting option. Pricing is quote-only and charged on annual candidate volume instead of per user, with all integrations included.
Vendor details
Canonical URL
https://goodtime.io
Category
Enterprise operations agent
Subcategory
AI interview scheduling & hiring coordination agent
Funding status
Established enterprise interview-scheduling vendor based in San Francisco; co-founded by CEO Ahryun Moon (named to Inc.'s 2026 Female Founders 500) and Chief Product Officer Jasper Sone. Customers include Databricks, Aon, HubSpot, HelloFresh, Priceline, Lyft, Talkdesk, and Glovo.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Connects to the applicant tracking system, calendar, video conferencing and messaging, with dedicated integration pages for Workday, Greenhouse, Lever, Ashby, iCIMS, Jobvite, SmartRecruiters and SAP SuccessFactors. Calendar access runs through the Microsoft Graph API under admin consent, with write scoped to events GoodTime itself created, and deep two-way ATS sync advances candidates through stages and writes scorecard exports back. Cori communicates over email, SMS, WhatsApp, Slack, Microsoft Teams, in-platform chat, voice and a branded candidate portal. All integrations are included in the subscription rather than sold as an add-on. A GoodTime MCP server, announced in July 2026 as a preview, connects external AI assistants to GoodTime data for natural-language query.
In practice
An interviewer cancels an hour before a five-person panel spanning three time zones. Cori finds the next best replacement, reschedules, and notifies everyone, instead of you rebuilding the whole slot by hand.
Your team still burns a third of its time on scheduling logistics. Cori runs the coordination layer end to end, booking panels, balancing interviewer load, and adapting in real time as calendars shift.
You worry an AI agent will act behind your back. Cori is built to be transparent. Every action it takes is visible, so recruiters stay informed and in control.
Sources & related URLs
Research sources
Agentic Index coverage score
9.0 / 14 capabilities · 64%
| Integrations & Tool Calling | Full |
|---|---|
|
GoodTime takes authenticated action in systems it does not own, across four classes. They are applicant tracking, with eight ATS integrations (Workday, Greenhouse, Lever, Ashby, iCIMS, Jobvite, SmartRecruiters, SAP SuccessFactors), calendar (Office 365 and Google), video conferencing, and messaging (email, SMS, WhatsApp, Slack, Microsoft Teams). On calendars, GoodTime uses the Office 365 Graph API to read and write interviewer calendars, with write limited to events GoodTime itself created, under an admin consent token plus an end user authorization token. Beyond that it runs deep two way ATS sync, advances candidates through stages, writes scorecard PDF exports into Workday, and sends candidate and interviewer communication through Slack, Teams, SMS and WhatsApp. Those are writes into outside systems, not reads or handoffs to a person. All integrations, including deep two way ATS sync, are included in the price instead of sold as an add-on. Sourcegoodtime.io/integrations, support.goodtime.io Office 365 integration guide and goodtime.io/pricing/hireread 2026-09-08 |
|
| Workflow Orchestration | Full |
|
Cori owns scheduling, screening, communication and workflow execution end to end. She screens and prioritizes candidates, schedules and reschedules interviews, coordinates interviewers and availability, converses with candidates in real time, sends updates and reminders, manages workflows and advances candidates through stages, and surfaces insights to fix bottlenecks. She acts without prompts and proactively detects conflicts. The participants are distinct and named. They are recruiters, hiring managers, interviewers and candidates, each with a different job in the sequence, plus the ATS, calendar and video systems the work runs through. This is coordination between parties, not one agent taking many steps in a single surface. Multi day panels across time zones with interviewer replacement mid flight require it. Workflow Automation, a separate feature offering configurable automation for every workflow, makes the sequence the customer's to define, and Cori escalates when something falls outside those defined workflows, which shows the boundary is a real configured object. Sourcegoodtime.io/products/hire/cori and /products/hire/recruiting-automationread 2026-09-08 |
|
| Knowledge Grounding & RAG | Full |
|
GoodTime grounds Cori in a persistent corpus of the customer's own recruiting data. It holds candidate profiles, job and interview plan details, interviewer records, availability, qualifications, training paths, tags and historical recruiting activity, kept current by continuous two way ATS and calendar synchronization instead of assembled per job. Profiles are AI summarized and feedback is consolidated. The corpus stands between runs and scales across an entire hiring organization far past any context window. It stays queryable. The July 2026 release ships Ask Cori about GoodTime, letting a user ask in plain English about interviews, interviewers, tags, scheduling or recruiting activity, and GoodTime MCP exposes the same corpus to external assistants for natural language query. Sourcegoodtime.io/blog/product-updates July 2026 and goodtime.io/products/hire/coriread 2026-09-08 |
|
| Human Oversight & Guardrails | Partial |
|
There is a configured boundary with automatic escalation, plus override, but no review and approve gate over Cori's ordinary actions. When a decision needs input or something falls outside the customer's defined workflows, Cori brings the team in, so the workflow definition is the edge and crossing it routes to a person instead of proceeding. Every action is visible and the team can step in, adjust or override at any time. On the natural language surface added in July 2026, Cori plans the work, validates every action, and keeps the user in control before anything is sent. Personally identifiable information is stripped before AI processing. Cori's normal path is still to act and then surface what it did. That validation applies to the natural language command surface, not to work the agent starts itself, and escalation happens on exceptions, not as a gate on every action. Sourcegoodtime.io/products/hire/cori and blog/product-updates July 2026read 2026-09-08 |
|
| Security, Identity & Governance | Full |
|
Attestations, with certificates, reports and dates, cover SOC 2 Type II (ISAE-3000), SOC 3, ISO 42001, HIPAA/HITECH Type 1, GDPR, CCPA and Data Privacy Framework, with an NYC Local Law 144 badge. The ISO 42001 certificate is viewable, GoodTime publishes dated penetration test certificates for GoodTime Hire, GoodTime Meet and the GoodTime API, and a Conveyor trust center holds the SOC 2 Type II report, HIPAA/HITECH Type 1 report, CAIQ Full, CAIQ Lite and SIG Lite under NDA. On access, GoodTime supports SAML 2.0 single sign on with a Super Admin SSO settings page, confirmed with Okta, OneLogin and Azure and open to any IdP. Identity provider deprovisioning propagates, so an account disabled at the domain level is disabled in GoodTime. A four level permission model (Super Admin, Admin, Member, No Permission) is set per user under Settings, Team, Permission. Least privilege is enforced on the calendar integration, where the Office 365 Graph scope writes only to events GoodTime itself created. Supporting controls include encryption at rest and in transit including backups, logical data separation, 72 hour breach notification with an incident report on request, defined and tested RTO and RPO, and 30 day data export on contract expiry. Sourcegoodtime.io/security and support.goodtime.io SAML/SSO and User Permission Levels articlesread 2026-09-08 |
|
| Observability & Auditability | Full |
|
Each action is recorded with its reasoning attached. Every action Cori takes is surfaced clearly, so the team always knows what is happening and why, and the team can step in, adjust or override at any time. This is a record of what the agent did, not an explanation of a score, covering actions, rationale and override. GoodTime does not show the inputs and assumptions behind each action, or whether a person accepted or changed it. Hiring analytics, dashboards and bottleneck detection give visibility into the customer's hiring process, not a record of the agent. Sourcegoodtime.io/products/hire/coriread 2026-09-08 |
|
| Memory & State Persistence | Not documented |
|
Cori adapts, but there is no memory store. Cori learns how the team hires and gets better over time, adapting to the team's preferences, priorities and workflows and learning from every interaction to constantly improve. This learning is absorbed into the product. There is no object holding taught criteria, no memory scope or stated lifetime, and no retention or deletion control over what Cori has learned as distinct from customer data. Candidate and interview state kept continuously synced with the ATS is the application's own data model. Deleting it deletes the interview record itself, which is what separates a data model from a memory layer. Sourcegoodtime.io/products/hire/coriread 2026-09-08 |
|
| Deployment & Data Residency | Partial |
|
Customers can have their data stored and processed on servers in the European Union. The option is stated in prose only. There is no region list, named customer environment or way to select a region, and GoodTime does not say how a customer elects EU hosting, whether as a contract term, a tier or a provisioning choice. GoodTime is hosted on Heroku and AWS, with the providers' physical data center and environmental controls, logically separated data and encrypted backups, and defined and tested RTO and RPO. Sourcegoodtime.io/securityread 2026-09-08 |
|
| Prebuilt Agents, Templates & Packs | Partial |
|
There is a product line, but not a set of agents to adopt. Nine features each have their own page, covering interview scheduling, applicant screening, workflow automation, SMS and WhatsApp messaging, candidate experience, interviewer experience, interviewer training, interview intelligence, and dashboards and analytics. They sit alongside nine industry solution pages and three use case pages. Cori is not a selectable unit, and GoodTime says so. Cori is not a separate product or add-on but the intelligence running across GoodTime, powering screening, scheduling, coordination and communication end to end. There is one agent and it is the platform, so it cannot be adopted, declined or chosen among. The nine features are a product line for a single engine, not complete working agents a customer selects among, and the industry pages are positioning. There is no browsable catalog, template library or named agent set. The 2025 Orchestra launch framed a digital workforce of multiple hiring agents. GoodTime has since collapsed that into one. Sourcegoodtime.io/products/hire/cori, /products/hire and site navigationread 2026-09-08 |
|
| Triggers & Channel Coverage | Full |
|
Cori is event driven by design. She does not wait for prompts, noticing what needs to happen and acting, proactively detecting conflicts and knowing when coordination is needed. She fires on inbound application flow, hiring stage changes, interviewer cancellations and availability shifts, screening and advancing candidates, scheduling and rescheduling, following up and triggering workflows automatically. Workflow Automation, a named feature offering configurable automation for every workflow, lets the customer configure the rules instead of receiving a fixed set, and the inbound side is a live two way sync, not a manual import. Channels include email, SMS and WhatsApp, Slack, Microsoft Teams, in platform chat, a branded candidate portal with self scheduling, and voice. A user can reach Cori by messaging her in Slack, sending a text, chatting in the platform or using voice. Sourcegoodtime.io/products/hire/cori and /products/hire/recruiting-automationread 2026-09-08 |
|
| Model Flexibility & Routing | Not documented |
|
No model choice is offered and no model provider is named for Cori. There is no model picker or admin model setting, provider list or routing description, or way to bring your own model. GoodTime MCP lets customers connect Claude, ChatGPT and other MCP compatible assistants to GoodTime data. It lets the customer choose which assistant reads GoodTime, and it says nothing about which model runs Cori. The subprocessor list and the Heroku and AWS hosting are infrastructure, not a model. Sourcegoodtime.io/products/hire/cori and /securityread 2026-09-08 |
|
| APIs, SDKs & MCP Extensibility | Partial |
|
GoodTime shows two signs of an extension surface but publishes no documentation for it. GoodTime MCP, announced in the July 2026 product release, connects AI assistants including Claude, ChatGPT and other MCP compatible tools directly to GoodTime data, so interviews, interviewers, tags and more can be queried in natural language, with an outside assistant driving GoodTime. A dated GoodTime API penetration test certificate sits alongside the Hire and Meet certificates, which shows a GoodTime API exists and has been tested. The API has been tested but has no public documentation. There is no developer portal, API reference or SDK, and the support center has integration guides for Office 365, Okta and OneLogin but no API documentation. GoodTime presents the MCP as a sneak peek, and that announcement still carries an unfilled placeholder, so its availability is unclear, not absent. Sourcegoodtime.io/blog/product-updates July 2026, goodtime.io/security and support.goodtime.ioread 2026-09-08 |
|
| Testing, Debugging & Optimization | Full |
|
A fairness evaluation runs against the customer's own deployment. The Bias Audit Report, a feature of the paid platform, lets a customer verify and continuously monitor that AI Candidate Match performs fairly across demographic groups including ethnicity, gender and other protected characteristics. The July 2026 release added reporting by country so fairness can be analyzed within each region, with the demographic profiles associated with the customer's results viewable. The customer invokes and reads it instead of requesting it as an engagement, it measures the agent's own matching decisions on the customer's own data, and results are comparable across groups and countries. GoodTime's NYC Local Law 144 badge, powered by Asenion, is a compliance disclosure about its product in general, not a test of the customer's deployment. Bottleneck detection and interviewer load optimization are live metrics on hiring operations. There is no sandbox before deployment, holdout or scored test case for the scheduling and coordination logic. Because pricing is contact only, GoodTime does not say which tier includes the Bias Audit Report. Sourcegoodtime.io/blog/product-updates July 2026read 2026-09-08 |
|
| Browser & Computer Use | Not documented |
|
There is no hosted or local browser, desktop session, or remote or local computer control. On a scheduling agent the closest feature can read like computer use. Cori operates interviewer calendars, books and moves meetings, and writes into the ATS and video conferencing. The mechanism is the Microsoft Graph API under an admin consent token with write scoped to events GoodTime created, a programmatic interface, not an agent driving a screen. Voice is a communication channel. Sourcegoodtime.io/products/hire/cori, /integrations and support.goodtime.io Office 365 guideread 2026-09-08 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Interview notes from GoodTime's AI Notetaker now flow into Greenhouse Scorecards. The notetaker takes word for word notes and sends the results straight back to Greenhouse for compliance and review, with no browser extension to install and no new scorecard templates to build.
Bears on: Integrations
View sourceGoodTime released a Model Context Protocol (MCP) integration that connects external AI assistants like Claude and ChatGPT directly to GoodTime data. The server integration allows users to query interviews, interviewers, and tags directly from their preferred AI clients.
Bears on: MCP / tool calling / API
View sourcePricing
Quote only
Annual candidate volume, not the number of users.
What is public
A pricing model without prices. Three structural facts are published. First, pricing is based on the number of candidates annually, not the number of users, and a customer can have as many users as needed. Second, all of the essential features to automate complex interview scheduling, engage candidates and gain hiring insights are included rather than tiered. Third, integrations, including deep two way ATS sync, are included rather than sold as an add on. The only route to a figure is a customized proposal via the request form.
Variable cost rationale
GoodTime charges on annual candidate volume, not per user, so spend follows hiring throughput, which fluctuates and does not depend on seat count. Users are unlimited, and every integration is included. The annual unit limits how fast the bill moves. Nothing is metered per message, per schedule or for AI usage. No volume bands, overage rule or cap are published, so the price at a given candidate volume is known only from a proposal.
Additional watchouts
The billing unit is annual candidate volume, so cost tracks hiring activity rather than headcount or seats. A good year for recruiting is a more expensive year for the tool, and a hiring freeze does not obviously reduce it. With no published bands, that exposure cannot be modeled before a sales conversation. On the other side, the model that ignores user count and the explicit inclusion of all integrations favor the customer.
Sales call required
Yes, required for paid access
Free / trial
None published
Lowest paid plan
Not published
Key ambiguities
The pricing model is published in unusual detail while every number is withheld, so a customer can see exactly how they would be charged and nothing about how much. Whether annual candidate volume is a committed tier or measured usage is not stated, and no overage rule appears.
Missing data
Every figure. No price, rate per candidate, volume band, minimum, floor or published tier appears. Contract length and billing frequency are unstated, as is whether annual candidate volume is committed in advance or trued up, and what happens on overage. That matters more here than on a per seat product, because the billing unit is the one a hiring team cannot forecast. GoodTime Meet is presented separately and its commercial relationship to Hire is not stated.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Abacum — FP&A platform for growing companies whose AI Modeler builds, edits…
- Adopt AI — Finance agents that categorize, reconcile and close the books inside…
- Aera Technology — Decision intelligence platform where an always on agent executes…
- Akro AI — On premise operational intelligence that automates document heavy…
- alfred_ — AI executive assistant that triages the inbox overnight and scores…
Alternatives to GoodTime
The closest documented capability profiles to GoodTime among enterprise operations agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Artos9.0 / 14Fuller documented coverage on Prebuilt Agents, Templates & Packs
- hireEZ8.0 / 14A lighter documented profile than GoodTime
- MarvelX9.0 / 14Fuller documented coverage on Human Oversight & Guardrails
- ketteQ9.5 / 14Fuller documented coverage on Human Oversight & Guardrails and Prebuilt Agents, Templates & Packs
- RedOwl8.5 / 14Fuller documented coverage on Human Oversight & Guardrails
- RegASK8.5 / 14Fuller documented coverage on Human Oversight & Guardrails
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded