{"id":165354,"date":"2026-11-02T13:16:48","date_gmt":"2026-11-02T06:16:48","guid":{"rendered":"https:\/\/bap-software.net\/?post_type=knowledge&p=165354"},"modified":"2026-11-02T13:24:53","modified_gmt":"2026-11-02T06:24:53","slug":"generative-ai-poc-to-production","status":"publish","type":"knowledge","link":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/","title":{"rendered":"From Generative AI PoC to Production: Evaluation Metrics and Go\/No-Go Criteria"},"content":{"rendered":"

<\/p>\n

\n

<\/p>\n

\n
\n

GENERATIVE AI \/ IMPLEMENTATION<\/p>\n

From Generative AI PoC to Production<\/p>\n

\n PoC & Validation<\/span>
\n \u2192<\/span>
\n Evaluation<\/span>
\n \u2192<\/span>
\n Production & Operations<\/span>\n <\/div>\n

Turning technical feasibility into sustainable business value<\/p>\n<\/p><\/div>

Figure 1. The journey from generative AI proof of concept to production deployment<\/figcaption><\/figure>\n

<\/p>\n

\n

GENERATIVE AI \/ KNOWLEDGE<\/p>\n

A successful AI proof of concept does not automatically mean a solution is ready for production.<\/p>\n

Many organizations can demonstrate generative AI capabilities in a controlled environment, yet struggle to move beyond experimentation. A prototype may produce promising responses, but decision-makers still need to understand whether it can deliver measurable business value, protect sensitive information, and operate reliably at scale.<\/p>\n

Moving from a generative AI proof of concept (PoC) to production requires more than model accuracy. Organizations must evaluate business impact, security, system integration, operational readiness, and ongoing costs. This guide explains the key PoC evaluation metrics, practical go\/no-go criteria, and a structured roadmap for enterprise deployment.<\/p>\n<\/p><\/div>\n

<\/p>\n

\n

What You’ll Learn<\/p>\n

    \n
  • How an AI proof of concept differs from a production-ready system<\/li>\n
  • Five essential metrics for evaluating generative AI PoCs<\/li>\n
  • How to make go\/no-go decisions based on measurable evidence<\/li>\n
  • How to translate PoC results into business value<\/li>\n
  • A five-step roadmap from validation to production operations<\/li>\n<\/ul><\/div>\n

    \"AI<\/p>\n

    \n <\/p>\n

    01. What Is a Generative AI PoC? PoC vs. Production<\/h2>\n

    A proof of concept (PoC) is a limited experiment designed to determine whether a proposed technology can solve a specific business problem. For generative AI, common PoC use cases include internal knowledge search, customer support, meeting summarization, document processing, and workflow assistance.<\/p>\n

    The objective is to validate technical feasibility and business relevance before committing to a larger implementation. Production deployment, however, introduces additional requirements such as reliability, access control, data governance, monitoring, and incident response.<\/p>\n

    \n\n\n\n\n\n\n\n\n
    Criteria<\/th>\nProof of Concept<\/th>\nProduction<\/th>\n<\/tr>\n<\/thead>\n
    Primary objective<\/td>\nValidate feasibility<\/td>\nDeliver ongoing business value<\/td>\n<\/tr>\n
    Scope<\/td>\nLimited tasks and users<\/td>\nReal users and operational workflows<\/td>\n<\/tr>\n
    Evaluation<\/td>\nAccuracy, feasibility, potential impact<\/td>\nQuality, security, reliability, ROI<\/td>\n<\/tr>\n
    System requirements<\/td>\nMinimal testing environment<\/td>\nMonitored and maintainable architecture<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n

    A PoC can meet its technical objectives without meeting production requirements. Establishing the distinction early helps teams define realistic acceptance criteria and avoid treating a promising demo as a finished enterprise solution.<\/p>\n

    <\/p>\n

    02. Five Essential Metrics for Evaluating an AI PoC<\/h2>\n

    Successful PoC evaluation requires more than asking whether an AI response sounds convincing. Organizations should measure the quality, usefulness, and operational implications of the system against clearly defined business requirements.<\/p>\n

    \n

    1. Answer Accuracy and Task Success Rate<\/p>\n

    Measure whether the system provides correct answers and completes intended tasks. Use representative test cases and reference answers, and classify errors such as incorrect responses, missing information, and unsuccessful actions.<\/p>\n

    2. Time Savings and Productivity<\/p>\n

    Compare the time required to complete tasks before and after introducing AI. Include the time employees spend reviewing, correcting, or approving AI-generated outputs.<\/p>\n

    3. Usability and User Acceptance<\/p>\n

    Assess whether employees can use the system effectively, verify supporting information, and incorporate the tool into existing workflows without unnecessary complexity.<\/p>\n

    4. Security and Data Governance<\/p>\n

    Review access permissions, confidential data handling, external model data transfers, logging policies, and compliance with applicable organizational requirements.<\/p>\n

    5. Cost and Operational Overhead<\/p>\n

    Estimate ongoing model or API charges, infrastructure costs, maintenance requirements, and the human effort needed to supervise the solution.<\/p>\n<\/p><\/div>\n

    \"AI<\/p>\n

    \n
    \n

    POC EVALUATION<\/p>\n

    Five Dimensions of AI PoC Success<\/p>\n

    \n 01 Accuracy<\/span>
    \n 02 Productivity<\/span>
    \n 03 Usability<\/span>
    \n 04 Security<\/span>
    \n 05 Operating Cost<\/span>\n <\/div>\n<\/p><\/div>
    Figure 2. Five key dimensions for evaluating generative AI proof-of-concept results<\/figcaption><\/figure>\n
    \n

    KEY POINT: Match evaluation metrics to the use case<\/p>\n

    For an internal knowledge assistant, factual accuracy and source traceability may be critical. For meeting summarization, coverage of key decisions and the time required for human corrections may matter more. Evaluation criteria should reflect the intended business outcome.<\/p>\n<\/p><\/div>\n

    <\/p>\n

    03. How to Make a Go\/No-Go Decision for Production<\/h2>\n

    At the end of a PoC, teams should compare actual results against the success criteria established before testing. The decision does not have to be a simple yes or no: a conditional approval may be appropriate when remaining issues are clearly defined and manageable.<\/p>\n

    \n\n\n\n\n\n\n\n
    Decision<\/th>\nTypical Situation<\/th>\nNext Action<\/th>\n<\/tr>\n<\/thead>\n
    Go<\/strong><\/td>\nRequired quality and business value demonstrated<\/td>\nProceed to production design and phased rollout<\/td>\n<\/tr>\n
    Conditional Go<\/strong><\/td>\nSome issues remain but can be addressed<\/td>\nResolve identified gaps and reassess<\/td>\n<\/tr>\n
    No-Go<\/strong><\/td>\nInsufficient value or unacceptable risks<\/td>\nRevisit requirements or the target use case<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n

    For example, a chatbot may achieve strong answer accuracy but still be unsuitable for deployment if it cannot meet data protection requirements. Likewise, a technically functional solution may not justify further investment if employees spend more time verifying outputs than they save.<\/p>\n

    Before approving production deployment, define acceptable thresholds for quality, business outcomes, security, and operational readiness, along with clear remediation plans for any unmet criteria.<\/p>\n

    <\/p>\n

    04. How to Measure the Business Value of an AI PoC<\/h2>\n

    To communicate PoC results to executives and business stakeholders, technical performance should be translated into measurable operational outcomes.<\/p>\n

    Consider a generative AI assistant designed to support internal employee inquiries. Beyond measuring answer accuracy, the organization should track the time employees spend responding to requests, including any review or correction work.<\/p>\n

    \n

    Illustrative Example: Internal Support Efficiency<\/p>\n

    Suppose an inquiry previously required 10 minutes to resolve. After introducing AI, the total time falls to 6 minutes, including human verification. This represents a saving of 4 minutes per inquiry.<\/p>\n

    At 1,000 inquiries per month, the potential time saving would be approximately 67 working hours per month.<\/p>\n

    Note: These figures are hypothetical and illustrate a calculation method. They do not represent actual BAP customer results.<\/p>\n<\/p><\/div>\n

    Time savings do not necessarily translate directly into financial profit. A realistic return-on-investment assessment should also consider development costs, AI service fees, infrastructure, maintenance, and human oversight.<\/p>\n

    Depending on the business objective, organizations may also measure consistency of service, employee experience, reduced information-search effort, or improved response quality.<\/p>\n

    <\/p>\n

    05. Five Steps to Move from AI PoC to Production<\/h2>\n

    Once the PoC has been evaluated, organizations need a practical transition plan that addresses technical architecture, business processes, and long-term operations.<\/p>\n

    \n

    STEP 01 | Review PoC Results and Open Issues<\/p>\n

    Document performance against evaluation criteria, recurring errors, user feedback, and unresolved technical or business risks.<\/p>\n

    STEP 02 | Define Production Requirements<\/p>\n

    Clarify user roles, expected usage, response-time requirements, access controls, data management policies, and incident-handling procedures.<\/p>\n

    STEP 03 | Design Enterprise System Integration<\/p>\n

    Plan integration with CRM, ERP, internal databases, or other business applications. Address APIs, authentication, data synchronization, and error handling.<\/p>\n

    STEP 04 | Launch a Controlled Production Rollout<\/p>\n

    Start with a limited user group or business unit, then monitor real-world performance, usability, and operational workload before expanding adoption.<\/p>\n

    STEP 05 | Monitor and Continuously Improve<\/p>\n

    Track output quality, usage, costs, and incidents. Refine the solution as business requirements, reference data, and AI technologies evolve.<\/p>\n<\/p><\/div>\n

    \"custom<\/p>\n

    \n
    \n

    IMPLEMENTATION ROADMAP<\/p>\n

    Five Steps to Production Readiness<\/p>\n

    \n Evaluate PoC<\/span>
    \n \u2192<\/span>
    \n Define Requirements<\/span>
    \n \u2192<\/span>
    \n Integrate Systems<\/span>
    \n \u2192<\/span>
    \n Phased Rollout<\/span>
    \n \u2192<\/span>
    \n Optimize<\/span>\n <\/div>\n<\/p><\/div>
    Figure 3. A five-step roadmap from PoC validation to ongoing production improvement<\/figcaption><\/figure>\n

    Generative AI systems may produce different outputs when models, prompts, or reference data change. Regular quality assessments using representative evaluation datasets should therefore continue after deployment.<\/p>\n

    <\/p>\n

    06. Three Ways to Avoid Common Production Deployment Mistakes<\/h2>\n

    1. Define Success Criteria Before Starting the PoC<\/strong>
    \n If evaluation criteria are established only after testing, results may be interpreted inconsistently. Set clear expectations for quality, business outcomes, and security before development begins.<\/p>\n

    2. Involve Business Users in Evaluation<\/strong>
    \n A solution that satisfies technical teams may still fail to fit daily operations. Include employees who will use the system in workflow testing, usability reviews, and feedback sessions.<\/p>\n

    3. Define Where Human Oversight Is Required<\/strong>
    \n Generative AI can produce inaccurate or unsupported information. For consequential decisions and business data updates, implement appropriate human review, approval, and escalation mechanisms.<\/p>\n

    \n

    Production deployment is the beginning, not the finish line<\/p>\n

    Long-term AI performance depends on monitoring, error analysis, data updates, security controls, and continuous improvement. Planning these responsibilities during development makes sustainable adoption more achievable.<\/p>\n<\/p><\/div>\n

    <\/p>\n

    07. Generative AI PoC and Production Development Services by BAP<\/h2>\n

    Successfully integrating generative AI into enterprise operations requires more than AI expertise. It also demands an understanding of business processes, system architecture, data security, and ongoing maintenance.<\/p>\n

    BAP combines capabilities in generative AI, retrieval-augmented generation (RAG), and AI agents with experience in web applications, enterprise software, and cloud development to support business-focused AI initiatives.<\/p>\n

    With development resources in Japan and Vietnam, BAP can support projects from requirements analysis and PoC development through evaluation, production implementation, system integration, and maintenance, depending on project needs.<\/p>\n

    <\/p>\n

    \n

    BAP SOLUTION JAPAN<\/p>\n

    Planning to Move Your Generative AI PoC into Production?<\/p>\n

    Whether you need to define PoC evaluation criteria, prepare a prototype for enterprise use, or integrate an AI solution with existing business systems, BAP can help you assess the next steps.<\/p>\n

    Contact BAP Solution Japan to discuss your generative AI PoC, production deployment, or system integration requirements.<\/p>\n<\/p><\/div>\n

    <\/p>\n

    Conclusion:<\/strong> Moving from a generative AI PoC to production requires a balanced assessment of answer quality, measurable business value, security, operational readiness, and cost. By defining clear acceptance criteria, deploying in stages, and establishing a continuous improvement process, organizations can turn promising AI experiments into practical business solutions.<\/p>\n<\/article>\n

    <\/p>","protected":false},"author":46,"featured_media":105801,"template":"","meta":{"_acf_changed":false},"tags":[],"blog-cat":[2058],"class_list":["post-165354","knowledge","type-knowledge","status-publish","has-post-thumbnail","hentry","blog-cat-technology"],"acf":[],"yoast_head":"\nGenerative AI PoC to Production: Evaluation & Roadmap<\/title>\n<meta name=\"description\" content=\"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\/no-go criteria, business ROI, security, and a five-step deployment roadmap.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"From Generative AI PoC to Production: Evaluation Metrics and Go\/No-Go Criteria\" \/>\n<meta property=\"og:description\" content=\"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\/no-go criteria, business ROI, security, and a five-step deployment roadmap.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/\" \/>\n<meta property=\"og:site_name\" content=\"Software development - offshore service | BAP Software\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/bap32\" \/>\n<meta property=\"article:modified_time\" content=\"2026-11-02T06:24:53+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/cdn.bap-software.net\/2025\/08\/05234958\/ai-tools-2-e1760430775435.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"717\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@bapsoftware\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"27 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/\",\"url\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/\",\"name\":\"Generative AI PoC to Production: Evaluation & Roadmap\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/bap-software.net\\\/knowledge\\\/generative-ai-poc-to-production\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/cdn.bap-software.net\\\/2025\\\/08\\\/05234958\\\/ai-tools-2-e1760430775435.jpg\",\"datePublished\":\"2026-11-02T06:16:48+00:00\",\"dateModified\":\"2026-11-02T06:24:53+00:00\",\"description\":\"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\\\/no-go criteria, business ROI, security, and a five-step deployment roadmap.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[[\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/\"]]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/#primaryimage\",\"url\":\"https:\\\/\\\/cdn.bap-software.net\\\/2025\\\/08\\\/05234958\\\/ai-tools-2-e1760430775435.jpg\",\"contentUrl\":\"https:\\\/\\\/cdn.bap-software.net\\\/2025\\\/08\\\/05234958\\\/ai-tools-2-e1760430775435.jpg\",\"width\":1000,\"height\":717,\"caption\":\"Ai tools\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/generative-ai-poc-to-production\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/bap-software.net\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Knowledge\",\"item\":\"https:\\\/\\\/bap-software.net\\\/en\\\/knowledge\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"\u751f\u6210AI\u306ePoC\u304b\u3089\u672c\u756a\u5c0e\u5165\u3078\u9032\u3080\u306b\u306f\uff1f\u8a55\u4fa1\u6307\u6a19\u3068\u5224\u65ad\u57fa\u6e96\u3092\u89e3\u8aac\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/bap-software.net\\\/en\\\/#website\",\"url\":\"https:\\\/\\\/bap-software.net\\\/en\\\/\",\"name\":\"Software development - offshore service | BAP Software\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/bap-software.net\\\/en\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Generative AI PoC to Production: Evaluation & Roadmap","description":"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\/no-go criteria, business ROI, security, and a five-step deployment roadmap.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/","og_locale":"en_US","og_type":"article","og_title":"From Generative AI PoC to Production: Evaluation Metrics and Go\/No-Go Criteria","og_description":"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\/no-go criteria, business ROI, security, and a five-step deployment roadmap.","og_url":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/","og_site_name":"Software development - offshore service | BAP Software","article_publisher":"https:\/\/www.facebook.com\/bap32","article_modified_time":"2026-11-02T06:24:53+00:00","og_image":[{"width":1000,"height":717,"url":"https:\/\/cdn.bap-software.net\/2025\/08\/05234958\/ai-tools-2-e1760430775435.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","twitter_site":"@bapsoftware","twitter_misc":{"Est. reading time":"27 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/","url":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/","name":"Generative AI PoC to Production: Evaluation & Roadmap","isPartOf":{"@id":"https:\/\/bap-software.net\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/#primaryimage"},"image":{"@id":"https:\/\/bap-software.net\/knowledge\/generative-ai-poc-to-production\/#primaryimage"},"thumbnailUrl":"https:\/\/cdn.bap-software.net\/2025\/08\/05234958\/ai-tools-2-e1760430775435.jpg","datePublished":"2026-11-02T06:16:48+00:00","dateModified":"2026-11-02T06:24:53+00:00","description":"Learn how to move a generative AI PoC into production. Explore evaluation metrics, go\/no-go criteria, business ROI, security, and a five-step deployment roadmap.","breadcrumb":{"@id":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":[["https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/"]]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/#primaryimage","url":"https:\/\/cdn.bap-software.net\/2025\/08\/05234958\/ai-tools-2-e1760430775435.jpg","contentUrl":"https:\/\/cdn.bap-software.net\/2025\/08\/05234958\/ai-tools-2-e1760430775435.jpg","width":1000,"height":717,"caption":"Ai tools"},{"@type":"BreadcrumbList","@id":"https:\/\/bap-software.net\/en\/knowledge\/generative-ai-poc-to-production\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/bap-software.net\/en\/"},{"@type":"ListItem","position":2,"name":"Knowledge","item":"https:\/\/bap-software.net\/en\/knowledge\/"},{"@type":"ListItem","position":3,"name":"\u751f\u6210AI\u306ePoC\u304b\u3089\u672c\u756a\u5c0e\u5165\u3078\u9032\u3080\u306b\u306f\uff1f\u8a55\u4fa1\u6307\u6a19\u3068\u5224\u65ad\u57fa\u6e96\u3092\u89e3\u8aac"}]},{"@type":"WebSite","@id":"https:\/\/bap-software.net\/en\/#website","url":"https:\/\/bap-software.net\/en\/","name":"Software development - offshore service | BAP Software","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/bap-software.net\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/knowledge\/165354","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/knowledge"}],"about":[{"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/types\/knowledge"}],"author":[{"embeddable":true,"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/users\/46"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/media\/105801"}],"wp:attachment":[{"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/media?parent=165354"}],"wp:term":[{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/tags?post=165354"},{"taxonomy":"blog-cat","embeddable":true,"href":"https:\/\/bap-software.net\/en\/wp-json\/wp\/v2\/blog-cat?post=165354"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}