[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間：2025-09-05 (世界標準時間)。"],[],[],null,["# Implementing Job search (v3)\n\nYou can implement the Cloud Talent Solution Job search API to leverage Google\nmachine learning (ML) technology with your existing search solution.\n\nImportant considerations\n------------------------\n\n1. The CTS Job search API is designed to be customizable, so you can configure\n the API according to your business needs. See [Best Practices](/talent-solution/job-search/v3/docs/best-practices) for a\n discussion of the parameters that you can adjust.\n\n2. Evaluate your capacity needs prior to implementing Job search. CTS should provide\n appropriate capacity to account for uploading jobs and/or companies, as well as handling daily operations\n such as job updates. Some important questions to consider when planning capacity:\n\n - How frequently are jobs Created, Updated and Deleted (CUD), and how often do job\n seekers receive job email alerts? There are 2 main use cases:\n\n 1. Capacity for the initial load or reload.\n 2. Capacity for normal operations.\n - How frequently are job seekers searching for jobs?\n\n 1. Capacity at peak.\n 2. Capacity for normal operations.\n\nUploading your data to the Job search API\n-----------------------------------------\n\n1. Before you can implement Job search, Cloud Talent Solution must be hooked up to your system.\n Follow the [quickstart](https://cloud.google.com/talent-solution/job-search/v3/docs/before-you-begin)\n guides to set up Cloud Talent Solution.\n\n2. Upload your jobs/companies data to the CTS Job search\n API. The API indexes your data alongside your existing database and uses pre-trained,\n built-in ML algorithms to return relevant search results.\n For most users, using the pre-trained model is a more than sufficient improvement.\n\n3. (Optional). You can further train the built-in CTS API algorithms\n using your own data.\n\n 1. To make efficient use of this option you need to identify 1 or 2 metrics that are most important to your business goals and track these throughout a period of training.\n 2. This phase generally takes weeks, depending on your system and the amount of uploaded data.\n 3. While training is underway, be sure to split your site traffic so that the performance of the CTS Job search API can be directly compared to your existing job search solution.\n\n Recommended setup:\n 1. CTS Job search API : 10%\n 2. CTS Job search API control group: 10%\n 3. Existing search solution: 80%\n\n At the conclusion of the training period you'll compare the performance of the trained CTS\n ML algorithm to your existing job search solution and make changes to the API configuration\n as needed.\n\nGo-live\n-------\n\nOnce the Job search API is integrated into your system and your site traffic is routed through it, the API\nwill begin returning relevant results to your users immediately. It's important to design\nyour integration solution in a fault-tolerant manner. That way, if for any reason the Job search API\nneeds to be tweaked you can route traffic back through your existing backend with no interruption\nto your users.\n\nLaunch checklist\n----------------\n\nFor a detailed checklist of tasks to complete during implementation, see our\n[launch checklist](/talent-solution/job-search/v3/docs/launch)."]]
