Jobo API

A one-sentence summary

The Jobo API is an infrastructure for job data designed for recruitment applications; it enables the unified search, synchronization, enhancement, and distribution of job records from numerous corporate recruitment sites and ATS systems.

What is the Jobo API?

Jobo is operated by JOBO AI Inc., and its main clients are job boards, HR technology platforms, recruitment research teams, as well as AI applications that require job data. It is not a chat tool for individuals to submit their resumes directly; rather, it is a developer service designed for use by product and engineering teams.

The platform claims to cover over 4 million active job listings, more than 200,000 company recruitment sites, and over 150 different ATS platforms. All job listings are organized in a unified structure, eliminating the need for developers to manage the data collection logic for systems such as Workday, Greenhouse, Lever, and iCIMS separately.

List of core capabilities

Job search API

The basic search interface supports full-text search as well as filtering by location, ATS channels, working mode, publication date, type of employment, experience level, skills, and annual salary range. The advanced search allows the use of multiple sets of keywords, location criteria, company specifications, and rules for including or excluding certain skills.

• Full-text search covers job titles, companies, and job descriptions.
• The location can be a city, state, country, or coordinates, and it will be parsed by the server.
• The working mode allows for selection between remote work, hybrid work, and on-site work.
• Employment types include full-time, part-time, contract, internship, and temporary positions.
• The experience levels range from beginner to intermediate and advanced, up to management and executive levels.
• Salary conditions are filtered based on the annual salary range in dollars.
• A single page returns up to 100 job listings, making it suitable for interactive result pages.

Batch synchronization of feeds

The Feed interface uses cursor-based pagination, returning up to 1000 job entries per batch; it is suitable for continuously synchronizing large volumes of job data into one’s own database. After an initial full import, incremental synchronization can be carried out based on the update time.

1. Create an account and generate a separate API key.
2. Determine the country, working mode, ATS channels, and time range for publication.
3. For the first request, no cursor is provided; this request retrieves the initial set of job listings along with the cursor for subsequent requests.
4. Repeat the request until no more data is returned.
5. Perform addition or update based on the Jobo position ID to avoid duplicate records.
6. Run incremental synchronization at regular intervals and record the time of the last successful execution.
7. The expired job interface is called weekly to clean up the publicly displayed listings.

Expired position management

The Expired Jobs interface is used to retrieve the IDs of jobs that have been removed or have expired; it supports cursor-based pagination and allows viewing of jobs from up to 7 days ago. It is free of charge, but customers still need to set up proper mechanisms to prevent invalid jobs from remaining on the public page for an extended period.

The terms of service require clients who list positions to update the list of expired positions at least once a week, and to remove those invalid positions in a manner that is commercially reasonable. Metadata such as position ID, title, company, location, and the user’s application history can be retained as needed for business purposes.

AI-enhanced fields

Jobo organizes the original job descriptions from different ATS systems into standardized fields, and uses AI to generate or extract structured information that is easy to search. The AI is used for data normalization and information extraction; it does not mean that the original employer has reviewed each field that is generated.

Job data field

A single record includes the job ID, title, company, full description, location, salary, type of employment, working pattern, experience level, ATS channel, publication date, expiration date, and application link. The company entity can also include a website, logo, and overview.

The page claims that each job posting includes more than 25 standardized fields. The completeness of the actual fields depends on the information provided by the recruitment site; fields such as salary, expiration date, visa requirements, and security permissions may be missing.

Outbound Feed and data pipelines

Teams that do not wish to develop their own pull tasks can use the Outbound pipeline to continuously insert data into their own databases or search infrastructure. The platform supports field mapping, filtering, transformation, AI-enhanced processing, incremental synchronization, deduplication, and retry of failed operations.

The product page describes continuous synchronization as updates made in increments of 15 minutes, while the Feed and Search pages indicate a refresh rate of around 1 hour. When making purchases, it is necessary to verify the actual freshness level for each specific product, account, and contract; the two different refresh rates should not be treated as one unified commitment.

MCP connects to the AI assistant

Jobo provides an MCP server for job searching, which allows compatible clients to access functions such as job searching, job details, and filtering tools via account authentication. It also offers a separate analysis MCP for performing aggregate queries on the entire dataset; the two serve different purposes.

Hosted MCP uses OAuth 2.1 for authentication, so there is no need to paste API keys in the chat client. Public packages also allow one to deploy their own gateway, but upstream requests still require authorization through a Jobo account; self-hosting does not entail access to offline data services.

Node.js and TypeScript clients

The official client is released as the jobo-enterprise package, and it includes functions for handling job feeds, searching, company information, and geocoding. It supports automatic pagination, incremental synchronization, type hints, error classification, as well as a limited number of retries for rate limiting issues and temporary service errors.

• Node.js 18 or a higher version is required.
• It can also be used in Bun, Deno, and modern browser environments.
• Authentication is carried out using an API key.
• It supports customizing timeout periods and request implementations.
• Fixed enums provide TypeScript autocompletion.
• The client code is licensed under the MIT license.
• Auto Apply is not currently encapsulated by this client.

Connectors and open-source code

The official GitHub organization has made available the n8n community nodes, WordPress plugins, Google Sheets extensions, MCP servers, the core connector components, and examples of Auto Apply. Most of these repositories are licensed under the MIT license, and they are labeled as one-way mirrors of internal monolithic repositories.

Making the client and connectors available publicly does not mean that the entire database of job listings, the data collection system, or the business platform are made open source. The service should be labeled as a closed-source SaaS product; only some of its components are available under the MIT license, not the whole system.

Current status of Auto Apply

The Auto Apply feature utilizes a unified interface to handle the multi-step application forms of various ATS systems, and it uses signature callbacks to return results regarding corrections, cancellations, and completions. This capability is still marked as upcoming, and its price has not yet been announced.

Jobo makes it clear that automatic applications do not use AI to generate responses for candidates; the person initiating the process must provide each answer manually. This design helps to reduce the occurrence of false information, but the party integrating this system still needs to obtain explicit permission from the candidates and verify the content submitted.

Package price

The pool of positions ranging from Lite to Scale is utilized by Search, Feed, Export, and Outbound; it is reset at the end of each month. Once the allocated amount is exceeded, charges are applied at the standard rate set by the wallet, with upgrades being calculated based on the remaining period.

Pay-as-you-go pricing and free tier

The system charges based on the actual number of job entries that are delivered, exported, or written, rather than on the number of requests made. If a single request returns ten job entries, ten units are charged; searches that yield no results do not incur any fee.

Limits of the Unlimited plan

Unlimited does not mean that all functions can be used indefinitely. It covers the Feed API, hosted Feeds, file export, and outbound writing, but job searches are still charged on a usage-based basis; geocoding and company data also have their own respective pricing.

The terms of the packages had their pricing adjusted on July 20, 2026; accounts created earlier may still retain the old packages and old pay-as-you-go rates. Existing customers should refer to the actual benefits listed in their account dashboard.

Billing, renewal, and refunds

• Subscriptions are prepaid on a monthly basis with automatic renewal.
• Cancelling must be completed before the start of the next billing cycle.
• The subscription fee for some months is usually not refundable.
• Unused wallet balance is not refundable, but it does not expire.
• The price does not include sales tax, value-added tax, or any other applicable taxes.
• When the balance is insufficient, operations that rely on the wallet will return an error indicating insufficient funds.
• The package’s included volume or unlimited usage option remains valid even when the balance is zero.

Access process

1. Register an account and receive an initial test balance.
2. Create separate keys for development, testing, and production environments in the console.
3. First, use a small number of searches to verify the fields, filtering criteria, and return structure.
4. Select Search or Feed based on interactive queries or full synchronization.
5. Implement mechanisms for removing duplicates, saving cursor states, retrying operations, and cleaning up expired job listings.
6. Response information and console monitoring include volume, wallet deductions, and rate limiting.
7. Before going live, check the data licensing, privacy requirements, and rules for public display.

Which users are it suitable for

• Recruitment platforms that need to quickly implement a job search function.
• Job websites that require the aggregation of data from multiple ATS systems.
• An AI application that helps candidates find jobs that match their skills.
• A recruitment data team that requires research on markets, skills, and salaries.
• Engineering teams that wish to synchronize job listings in databases or search engines.
• Operators who use n8n, WordPress, or tables to create lightweight workflows.
• Agent developers who need AI assistants to search for real job opportunities.

Product advantages

• A unified interface covers over 150 ATS platforms.
• It takes into account interactive searching, batch Feed, file export, and proactive push notifications.
• Fields such as job summary, skills, experience, and salary have been standardized.
• Charging is based on the volume of work performed per position, making it easy to estimate the costs.
• New accounts come with a small free test balance.
• It also offers REST, MCP, SDKs, and low-code connectors.
• The public interface for expired positions facilitates the maintenance of their relevance.
• Data licensing allows for the long-term storage and display of delivery records within one’s own products.

Main limitations

• The data comes from public recruitment websites, and its completeness and accuracy are not guaranteed.
• AI summaries and structured fields may misinterpret the original job description.
• Some positions lack information on salary, deadline, or visa requirements.
• Both searching and batch synchronization are subject to rate limits set by account and endpoint.
• Unlimited does not include unlimited job searches.
• It is not allowed to resell the original data or create APIs for competitive job data.
• General AI models intended for third parties cannot be trained using delivery data.
• Auto Apply has not yet been officially launched, and its price has not been announced.
• The synchronization frequency varies across different product specifications and must be confirmed through a contract.

Data licensing

Customers are granted a non-exclusive, global, and permanent license to use the data related to the positions that have been delivered; they can cache this data, index it, display it, remove duplicates from it, and sort it. Such data can also be used in matching, classification, and recommendation algorithms within their own products. Even after the account is terminated, the data obtained in compliance with relevant regulations can usually still be used.

Currently, the publicly displayed job listings do not require the inclusion of the Jobo designation, but the platform reserves the right to impose such a requirement after giving 30 days’ notice. The employer’s name, logo, and the description of the job may still involve third-party rights, and clients are required to respond to requests for correction or removal.

Privacy and security

Registration collects the name, email address, and an optional company name; payments are processed by Stripe, with Jobo only receiving essential information such as the payment status and the last four digits of the card number. API request logs record the endpoint, time, status, and key identifier, and are typically retained for 30 days.

The platform claims that both transmitted data and static data are encrypted, and it sets a monthly availability target of 99.9% for production endpoints. Integrators should still store keys in a server-side secret management system, create keys for different environments, and monitor any abnormal calls.

Frequently Asked Questions

Is the Jobo API meant for job seekers?

It is primarily a data service for job positions aimed at developers and those involved in recruitment. Individuals usually make use of it indirectly by accessing Jobo’s job website or its AI assistant.

Is there a free version?

New accounts receive a starting balance of $5 and do not require a credit card; there is no permanent, unlimited free plan. Once the balance is used up, it is necessary to top up the account or purchase a monthly plan.

How is pricing calculated for each request?

Charging is based on the number of job entries that are returned, exported, or written, not on the number of calls made. Once the available quantity is exhausted, a fee of $3 per 1,000 entries will be deducted from the wallet.

Does Unlimited include job searching?

It does not include unlimited searching. Unlimited coverage is available for Feeds, hosted Feeds, exports, and Outbound; Search remains subject to pay-per-use pricing.

Is the Jobo API open source?

The core platform and the services related to job data are not open-source. The official version of MCP, connectors, n8n nodes, plugins, and the Node.js client are available under the MIT license.

Can the data be resold to other companies?

It cannot be resold, relicensed, or provided in bulk dump form in its basic original form. It can be used to display job listings to users within own end-product applications, along with functions for searching, matching, and recommending jobs.

Can AI be trained using job data?

It can be used for matching, sorting, and classification models within proprietary products, but it cannot be used to train general AI models that are provided to third parties.

Is Auto Apply already available?

It has not been officially launched yet, and its price has not been announced. The page offers only a preview of its capabilities and example integrations; it cannot be used as a ready-to-use production interface for purchase.

Summary

The Jobo API is suitable for recruitment platforms that require reliable job searching, full synchronization, AI-driven field enhancement, and multi-target job distribution. REST, MCP, Node.js clients, and connectors constitute a comprehensive developer ecosystem.

When selecting a package, it is necessary to calculate the costs based on the actual number of positions delivered each month, the Search ratio, and the synchronization method; at the same time, measures such as the removal of expired positions, manual verification of fields, key management, and data licensing requirements must be taken into account. The boundaries of the Unlimited plan, the synchronization frequency, and the Auto Apply status need to be confirmed in advance before going live.