<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Raghu Nandy</title>
    <description>The latest articles on DEV Community by Raghu Nandy (@raghunandy).</description>
    <link>https://dev.to/raghunandy</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F74870%2Ff8747ca3-7ba1-4873-b621-e096c24f7c66.png</url>
      <title>DEV Community: Raghu Nandy</title>
      <link>https://dev.to/raghunandy</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/raghunandy"/>
    <language>en</language>
    <item>
      <title>Sustainable AI Needs Better Service Models: Introducing ORSi Spec</title>
      <dc:creator>Raghu Nandy</dc:creator>
      <pubDate>Wed, 18 Nov 2026 07:33:58 +0000</pubDate>
      <link>https://dev.to/raghunandy/sustainable-ai-needs-better-service-models-introducing-orsi-spec-eda</link>
      <guid>https://dev.to/raghunandy/sustainable-ai-needs-better-service-models-introducing-orsi-spec-eda</guid>
      <description>&lt;p&gt;&lt;em&gt;Exploring independent AI providers, reusable workflows, Results as a Service (RaaS), modern jobs, and long-term ecosystem growth.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fmd8qhrvfpz7l38kov7gx.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fmd8qhrvfpz7l38kov7gx.png" alt="Concept illustration of a silver AI box with four portable USB devices on a desk, ORSi Spec text, and a GitHub repository link." width="800" height="600"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Concept illustration of one possible device-associated AI service application—not a shipping product or certification.&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  AI is growing. Are our service models keeping up?
&lt;/h2&gt;

&lt;p&gt;AI developers, inference companies, and entrepreneurs are building increasingly capable systems.&lt;/p&gt;

&lt;p&gt;But growth also brings questions about computing resources, service costs, customer expectations, and long-term business strategies.&lt;/p&gt;

&lt;p&gt;A sustainable AI ecosystem needs more than capable models. It needs practical ways for independent providers and developers to offer services under clear arrangements.&lt;/p&gt;

&lt;p&gt;That is one motivation behind &lt;strong&gt;ORSi Spec — OperatoR System and Intelligence&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;For AI Development, Distribution &amp;amp; Sustainability.&lt;/em&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What is ORSi Spec?
&lt;/h2&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fmfvvikr74clhf2o21hhb.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fmfvvikr74clhf2o21hhb.png" alt="Notebook infographic outlining the ORSi review draft: common service-access concepts, telecom-inspired examples, device-associated access, local/cloud/hybrid AI, provider terms, and a four-stage service-access diagram." width="800" height="1000"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;ORSi is a proposed common foundation for describing AI service access while preserving each provider's independence.&lt;/p&gt;

&lt;p&gt;The first review draft, &lt;strong&gt;ORSi-01 — Device-Associated AI Service Entitlements&lt;/strong&gt;, focuses on a small set of shared concepts:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Issuer:&lt;/strong&gt; Who authorizes the AI service?&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Service:&lt;/strong&gt; What does the customer receive?&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Device scope:&lt;/strong&gt; Which enrolled device or devices are covered?&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Validity:&lt;/strong&gt; When does access begin and end?&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Terms:&lt;/strong&gt; What usage conditions and commercial policies apply?&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The aim is not to create another compulsory marketplace, payment gateway, or model runtime.&lt;/p&gt;

&lt;p&gt;It is to explore whether these basic concepts can be expressed consistently while leaving actual business arrangements to independent providers.&lt;/p&gt;

&lt;h2&gt;
  
  
  A practical example
&lt;/h2&gt;

&lt;p&gt;Imagine an independent developer offering an AI invoice-processing workflow.&lt;/p&gt;

&lt;p&gt;One company supplies model inference. Another delivers the workflow and customer support. The customer subscribes to the service.&lt;/p&gt;

&lt;p&gt;Some tasks may require fresh inference. Others might be handled by an existing reviewed routine, depending on the inputs and conditions.&lt;/p&gt;

&lt;p&gt;That introduces several important questions:&lt;/p&gt;

&lt;p&gt;Who grants access? Which tasks are included? How long is the service valid? Which company charges the customer? How are actual inference charges distinguished from other service fees?&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;ORSi-01 proposes shared entitlement concepts for describing such arrangements.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;It does not prescribe the workflow implementation or a particular pricing structure.&lt;/p&gt;

&lt;h2&gt;
  
  
  SaaS, RaaS, and new AI business opportunities
&lt;/h2&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fxjbgvhijpgaiihs8c4lj.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fxjbgvhijpgaiihs8c4lj.png" alt="Illustrative SaaS versus RaaS comparison: access-based and result-oriented offerings, potential customer-retention considerations, and areas where ORSi could support provider-defined service strategies." width="800" height="890"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Concept comparison only: RaaS is not proven to produce lower churn than SaaS in every setting.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;Many traditional SaaS offerings charge customers for access, seats, or subscriptions.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Results as a Service (RaaS)&lt;/strong&gt; explores another possibility: offering services around clearly defined, measurable outcomes.&lt;/p&gt;

&lt;p&gt;For example, an entrepreneur might develop a service for processing documents, completing business workflows, or resolving support requests.&lt;/p&gt;

&lt;p&gt;Rather than focusing only on access to software, the provider could define an offering around agreed results.&lt;/p&gt;

&lt;p&gt;This may encourage:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Workflow innovation:&lt;/strong&gt; Developers compete on useful solutions, not simply model access.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Customer retention:&lt;/strong&gt; Consistent, meaningful outcomes may give customers stronger reasons to remain, potentially reducing avoidable churn.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Modern jobs:&lt;/strong&gt; Opportunities may emerge for workflow designers, AI integration specialists, independent operators, and service professionals.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Entrepreneurship:&lt;/strong&gt; Freelancers and small businesses can explore recurring AI services rather than relying entirely on one-time projects.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Provider flexibility:&lt;/strong&gt; Inference companies can continue defining their own usage and commercial terms.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;RaaS is not automatically better than SaaS. Outcomes must be measurable, attribution needs to be credible, and charges must follow agreed terms. Revenue may still vary with the volume of outcomes delivered.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The important point is choice.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;ORSi does not require RaaS, SaaS, or any particular combination of the two.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why independence matters
&lt;/h2&gt;

&lt;p&gt;A useful AI ecosystem should leave room for model developers, inference providers, independent entrepreneurs, freelancers, businesses, and customers.&lt;/p&gt;

&lt;p&gt;Different participants have different constraints, resources, and commercial strategies.&lt;/p&gt;

&lt;p&gt;A shared specification should not automatically transfer pricing control, customer accounts, or governance authority to one industry.&lt;/p&gt;

&lt;p&gt;That leads to our core principle:&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;ORSi is open to everyone, owned by no industry sector.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;And our strategic position:&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;ORSi is not a business model. It is a proposed foundation that allows different AI business strategies to coexist and evolve.&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  What does sustainable AI growth mean?
&lt;/h2&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fzq67hj8xrdvnxz801m7g.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fzq67hj8xrdvnxz801m7g.png" alt="Notebook illustration of ORSi's proposed contribution to provider freedom, reusable workflows, clear customer access, and independent services, with a sunrise scene symbolizing possible new career and entrepreneurship opportunities." width="800" height="889"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Illustrative future opportunities, not measured ORSi outcomes or guaranteed job creation.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;AI adoption can expand rapidly, while infrastructure and service delivery continue to require significant resources.&lt;/p&gt;

&lt;p&gt;Inference providers need flexibility to manage capacity, usage, and independently agreed commercial arrangements.&lt;/p&gt;

&lt;p&gt;AI entrepreneurs and freelancers need opportunities to create valuable, reusable services that can grow alongside customer demand.&lt;/p&gt;

&lt;p&gt;Customers need understandable terms and services that consistently deliver value.&lt;/p&gt;

&lt;p&gt;A healthy ecosystem should encourage innovation across these roles.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Sustainable AI growth is not just about scaling technology. It is about creating conditions in which different participants can continue innovating together.&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;ORSi explores one possible foundation for supporting that direction.&lt;/p&gt;

&lt;h2&gt;
  
  
  What is ready today?
&lt;/h2&gt;

&lt;p&gt;ORSi-01 is publicly available as a &lt;strong&gt;review draft&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;It is not a ratified standard, certified implementation, or proof that the proposed approach improves business outcomes.&lt;/p&gt;

&lt;p&gt;The draft does not define a complete inference protocol, universal billing system, hardware authentication mechanism, or outcome-verification standard.&lt;/p&gt;

&lt;p&gt;These boundaries are deliberate.&lt;/p&gt;

&lt;p&gt;The goal is to start with a small proposal and let technical feedback identify what is genuinely necessary.&lt;/p&gt;

&lt;p&gt;Future extensions should emerge from demonstrated needs rather than adding unnecessary complexity to the first profile.&lt;/p&gt;

&lt;h2&gt;
  
  
  What would make this useful?
&lt;/h2&gt;

&lt;p&gt;I'm especially interested in hearing from AI developers, inference providers, entrepreneurs, and technical reviewers.&lt;/p&gt;

&lt;p&gt;Three questions are worth exploring:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Are these the right minimum concepts for describing AI service access?&lt;/li&gt;
&lt;li&gt;Which parts should remain entirely provider-defined?&lt;/li&gt;
&lt;li&gt;Could result-oriented AI workflows support better customer retention, modern jobs, and more sustainable service strategies?&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;&lt;strong&gt;Review ORSi Spec on GitHub:&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href="https://github.com/orsi-spec-project/operator-system-and-intelligence-spec" rel="noopener noreferrer"&gt;https://github.com/orsi-spec-project/operator-system-and-intelligence-spec&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Technical criticism, alternative approaches, and practical examples are welcome.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Better AI service models should create room for independent innovation and long-term ecosystem growth.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>saas</category>
      <category>raas</category>
      <category>discuss</category>
    </item>
  </channel>
</rss>
