Add CLAUDE guide; refocus site for Azure & production

CLAUDE.md

@@ -0,0 +1,177 @@
+# CLAUDE.md
+
+This file gives Claude (and any other AI assistant) the working context required to safely edit the Umplify website. The site is a Jekyll project deployed via GitHub Pages at [https://umplify.com](https://umplify.com).
+
+## What this site is
+
+Umplify is a Toronto-based AI transformation and Azure cloud platform engineering partner. The website is a marketing and lead-generation site for software-led businesses, ISVs, scale-ups, and enterprise teams in Canada and the United States.
+
+Primary visitor goals (in priority order):
+
+1. Book a free discovery call (`/contact/`).
+2. Understand the AI transformation and Azure platform service offering.
+3. Read insight content on the blog at `/blog/`.
+
+## Tech stack
+
+- **Static site generator:** Jekyll, deployed via GitHub Pages.
+- **Theme:** [`mmistakes/minimal-mistakes`](https://github.com/mmistakes/minimal-mistakes) v4.24.0 via `remote_theme` (do not vendor the theme).
+- **Markdown:** kramdown with GFM input.
+- **Plugins (must stay GitHub Pages compatible):**
+  - `jekyll-paginate`
+  - `jekyll-sitemap`
+  - `jekyll-gist`
+  - `jekyll-feed`
+  - `jekyll-include-cache`
+  - `jekyll-redirect-from`
+- **Hosting:** GitHub Pages with custom domain via `CNAME`.
+- **Analytics:** Google gtag (`G-WSEKFYVPNE`) configured in `_config.yml`.
+
+If a change requires a plugin not on the GitHub Pages allow-list, do not add it. Find another approach.
+
+## Repository layout
+
+```
+_config.yml              Site config (title, plugins, defaults, analytics)
+_data/
+  navigation.yml         Top nav order and labels
+  ui-text.yml            Theme UI strings (do not edit lightly)
+_includes/               Theme overrides and partials
+  head/custom.html       Custom head: favicon, OG tags, JSON-LD schemas
+  masthead.html          Top nav (overrides theme default)
+  footer.html, etc.
+_layouts/                Theme layouts (single, splash, home, archive, etc.)
+_pages/                  All non-blog pages (services, home, about, contact, faq)
+_posts/                  Blog articles. Filename pattern: YYYY-MM-DD-slug.md
+_sass/                   Theme SCSS overrides
+assets/
+  images/
+    revamp/              Current hero/diagram SVGs
+    splash/              Feature-row icons
+    logos/               Brand logos
+robots.txt               Allows all, points to sitemap.xml
+CNAME                    umplify.com
+```
+
+## Content authoring rules
+
+### Brand voice
+
+- **Plain, senior, production-first.** The reader is a CTO, VP Engineering, founder, or senior architect. Speak in their language.
+- **No fluff.** Every sentence either establishes credibility or drives a decision.
+- **No em dashes (—).** Use commas, periods, semicolons, parentheses, or restructure the sentence. (See `memory/feedback_no_em_dashes.md` in the user's memory.)
+- **Canadian English** spellings where they differ (e.g., "modernize" is fine, "centre" not "center" only in proper nouns; the site uses US-style "modernize/optimize").
+- **Avoid hype words:** "revolutionary", "cutting-edge", "10x", "game-changer", "transformative" without proof.
+- **Outcomes before features.** Lead with what the reader gets, then explain how.
+- **CTA ladder.** Every important page should end with at least one primary CTA (`Book a free discovery call`) and one secondary (e.g., `How we engage`, `Read the FAQ`).
+
+### Page front matter conventions
+
+Service pages should look like this:
+
+```yaml
+---
+title: "Service Name"
+permalink: /service-slug/
+layout: single
+author_profile: false
+service_schema: true        # renders Schema.org Service JSON-LD via _includes/head/custom.html
+excerpt: "One sentence used for meta description and OG/Twitter card."
+header:
+  overlay_color: "#08142C"
+  overlay_filter: "0.22"
+  overlay_image: /assets/images/revamp/hero-ai.svg   # AI pages
+  # or /assets/images/revamp/hero-cloud.svg          # platform/Azure pages
+redirect_from:              # optional, requires jekyll-redirect-from
+  - /old-url/
+---
+```
+
+Landing pages (home, ai-transformation, cloud-solutions) use `layout: splash` and feature rows. See `_pages/home.md` for the canonical pattern.
+
+### Permalinks
+
+Service pages use top-level slugs (`/agentic-workflow-design/`). Legacy URLs under `/cloud-solutions/...` are kept as `redirect_to:` stubs for SEO continuity. Do not break existing permalinks. If a page is retired, replace its content with a stub:
+
+```yaml
+---
+title: "..."
+permalink: /old/url/
+sitemap: false
+redirect_to: /new/url/
+---
+```
+
+### Internal linking
+
+Every service page should:
+
+1. Link to at least three related service pages.
+2. Link to `/contact/`.
+3. Link to `/how-we-engage/` if appropriate.
+
+Use markdown link syntax with leading slash: `[Label](/path/)`.
+
+### Images
+
+- Hero/header images live in `assets/images/revamp/`.
+- Feature row icons live in `assets/images/splash/`.
+- Always include alt text on `![alt](path)`.
+- SVG is preferred. Avoid raster heroes when an SVG exists.
+- The site `og_image` is `/assets/images/revamp/hero-ai.svg`. Some social platforms reject SVG OG images; prefer setting a page-specific `header.overlay_image` so the head template emits a usable URL.
+
+### SEO and structured data
+
+`_includes/head/custom.html` already emits:
+
+- `Organization` JSON-LD on every page.
+- `Service` JSON-LD when a page has `service_schema: true` in its front matter.
+- `FAQPage` JSON-LD on `/faq/` (kept in sync with `_pages/faq.md`).
+- `BlogPosting` JSON-LD on blog posts.
+
+If you add a new FAQ question on `_pages/faq.md`, also update the `FAQPage` JSON-LD block in `_includes/head/custom.html`. They must stay synchronized.
+
+If you add a new service page, set `service_schema: true` to get Service JSON-LD for free.
+
+### Navigation
+
+Top nav is driven by `_data/navigation.yml`. Keep the list short (target eight items or fewer). New service detail pages should be cross-linked from the relevant landing page (`/ai-transformation/` or `/cloud-solutions/`), not added to the top nav.
+
+## Blog posts
+
+Blog posts live in `_posts/` with filenames `YYYY-MM-DD-slug.md`. They are excluded from the audit process Claude runs on marketing pages. **Do not modify blog articles unless explicitly asked.** When writing new posts:
+
+- No em dashes (this is a project-wide rule, not blog-only).
+- Use the same kramdown/GFM conventions as existing posts.
+- Posts inherit `layout: single` and other defaults from `_config.yml`.
+
+## Local development
+
+```bash
+bundle install
+bundle exec jekyll serve
+# open http://127.0.0.1:4000/
+```
+
+Built output goes to `_site/`. **Never commit `_site/`.**
+
+## Things to avoid
+
+1. **Em dashes anywhere in site content.**
+2. Adding plugins outside the GitHub Pages allow-list.
+3. Editing files inside the `mmistakes/minimal-mistakes` theme directly. Override via `_includes/`, `_layouts/`, or `_sass/` instead.
+4. Changing existing permalinks without a `redirect_from:` (or `redirect_to:` stub) covering the old URL.
+5. Committing build artifacts (`_site/`, `.jekyll-cache/`, `.sass-cache/`).
+6. Hard-coding the site URL. Use `{{ site.url }}` or `| absolute_url` / `| relative_url` filters.
+
+## Useful tasks for Claude
+
+- **Add a new service page.** Copy the front matter pattern above, set `service_schema: true`, write the body, cross-link from `_pages/ai-transformation.md` or `_pages/cloud-solutions.md`, and link related pages from inside the new page.
+- **Retire a page.** Replace its content with the `redirect_to:` stub pattern shown above and verify nothing else in the repo links to it (`grep -r "/old-url/" _pages _data _includes`).
+- **Add a new FAQ question.** Edit both `_pages/faq.md` and the `FAQPage` JSON-LD block inside `_includes/head/custom.html`.
+- **Update the brand description.** Update `_config.yml` (`title`, `description`, `subtitle`) and re-check `_includes/head/custom.html` `Organization` JSON-LD if structural details change.
+- **Audit content for em dashes before commit.** `grep -rn "—" _pages _data _includes _config.yml _posts`. The result should be empty.
+
+## Contact for questions
+
+Repository owner: Arash Sabet (`arash.sabet@umplify.com`). The site reflects positioning that has been deliberately chosen; large strategic shifts (positioning, pricing language, ICP) should be confirmed before being shipped.

_config.yml

@@ -11,11 +11,11 @@ minimal_mistakes_skin    : "default"
 
 # Site Settings
 locale                   : "en-US"
-title                    : "Umplify | AI Transformation and Cloud Platform Engineering"
+title                    : "Umplify | AI Transformation and Azure Cloud Platform Engineering"
 title_separator          : "|"
-subtitle                 : "Your AI and Cloud transformation partner"
+subtitle                 : "AI-ready systems. Production-grade Azure platforms."
 name                     : "Umplify Technologies Inc."
-description              : "Umplify helps businesses in the Greater Toronto Area and across Canada become AI-ready through AI transformation services, enterprise AI integration, agentic workflows, and Azure cloud platform engineering."
+description              : "Umplify is a Toronto-based AI transformation and Azure cloud platform engineering partner. We design agentic workflows, enterprise AI integrations, private knowledge systems, and production-grade Azure architecture for software-led businesses across Canada."
 url                      : "https://umplify.com"
 baseurl                  :
 repository               : "Umplify/umplify"
@@ -180,13 +180,15 @@ plugins:
   - jekyll-gist
   - jekyll-feed
   - jekyll-include-cache
+  - jekyll-redirect-from
 
 whitelist:
   - jekyll-paginate
   - jekyll-sitemap
   - jekyll-gist
   - jekyll-feed
   - jekyll-include-cache
+  - jekyll-redirect-from
 
 category_archive:
   type: liquid

_data/navigation.yml

@@ -3,8 +3,10 @@ main:
     url: /
   - title: "AI Transformation"
     url: /ai-transformation/
-  - title: "Cloud Platform Engineering"
+  - title: "Azure Platform"
     url: /cloud-solutions/
+  - title: "How We Engage"
+    url: /how-we-engage/
   - title: "Blog"
     url: /blog/
   - title: "FAQ"

_includes/head/custom.html

@@ -25,25 +25,86 @@
   "@context": "https://schema.org",
   "@type": "Organization",
   "name": "Umplify Technologies Inc.",
+  "alternateName": "Umplify",
   "url": "{{ site.url }}",
   "logo": "{{ absolute_logo }}",
   "description": "{{ site.description | escape }}",
+  "foundingLocation": {
+    "@type": "Place",
+    "name": "Toronto, Ontario, Canada"
+  },
   "areaServed": [
     {"@type": "AdministrativeArea", "name": "Greater Toronto Area"},
-    {"@type": "Country", "name": "Canada"}
+    {"@type": "Country", "name": "Canada"},
+    {"@type": "Country", "name": "United States"}
+  ],
+  "knowsAbout": [
+    "AI Transformation",
+    "Agentic Workflows",
+    "Enterprise AI Integration",
+    "Microsoft Azure",
+    "Azure OpenAI",
+    "Cloud Platform Engineering",
+    "SaaS Architecture",
+    "DevOps",
+    "Infrastructure as Code"
   ],
   "address": {
     "@type": "PostalAddress",
     "addressLocality": "Toronto",
     "addressRegion": "ON",
     "addressCountry": "CA"
   },
+  "contactPoint": [
+    {
+      "@type": "ContactPoint",
+      "contactType": "sales",
+      "email": "sales@umplify.com",
+      "areaServed": ["CA", "US"],
+      "availableLanguage": "en"
+    },
+    {
+      "@type": "ContactPoint",
+      "contactType": "customer support",
+      "email": "support@umplify.com",
+      "areaServed": ["CA", "US"],
+      "availableLanguage": "en"
+    }
+  ],
   "sameAs": [
     "https://github.com/Umplify/umplify"
   ]
 }
 </script>
 
+{% if page.service_schema %}
+<script type="application/ld+json">
+{
+  "@context": "https://schema.org",
+  "@type": "Service",
+  "serviceType": {{ page.title | jsonify }},
+  "name": {{ page.title | jsonify }},
+  "description": {{ seo_description | strip_html | strip_newlines | jsonify }},
+  "provider": {
+    "@type": "Organization",
+    "name": "Umplify Technologies Inc.",
+    "url": "{{ site.url }}"
+  },
+  "areaServed": [
+    {"@type": "AdministrativeArea", "name": "Greater Toronto Area"},
+    {"@type": "Country", "name": "Canada"},
+    {"@type": "Country", "name": "United States"}
+  ],
+  "audience": {
+    "@type": "BusinessAudience",
+    "audienceType": "Software-led businesses, startups, and enterprise teams"
+  },
+  "url": "{{ page.url | absolute_url }}",
+  "image": "{{ absolute_seo_image }}"
+}
+</script>
+{% endif %}
+
 {% if page.url == '/faq/' %}
 <script type="application/ld+json">
 {
@@ -55,23 +116,63 @@
       "name": "What does it mean for a business to become AI-ready?",
       "acceptedAnswer": {
         "@type": "Answer",
-        "text": "An AI-ready business has the workflows, systems, knowledge access, permissions, governance, and platform architecture needed to support useful AI implementation."
+        "text": "An AI-ready business has the workflows, systems, knowledge access, permissions, governance, and platform architecture needed to support useful AI implementation in production rather than only in pilots."
       }
     },
     {
       "@type": "Question",
       "name": "What are agentic workflows?",
       "acceptedAnswer": {
         "@type": "Answer",
-        "text": "Agentic workflows are business workflows where AI can retrieve information, support decisions, coordinate tasks, and interact with tools or business systems within defined controls."
+        "text": "Agentic workflows are business workflows where AI can retrieve information, support decisions, coordinate multi-step tasks, and interact with tools or business systems within defined human-in-the-loop controls."
+      }
+    },
+    {
+      "@type": "Question",
+      "name": "Where does MCP-oriented integration fit?",
+      "acceptedAnswer": {
+        "@type": "Answer",
+        "text": "MCP-oriented integration matters when AI systems need structured, secure access to internal tools, data, and actions. It shapes how access layers, context handling, and system interoperability are designed for production use cases."
       }
     },
     {
       "@type": "Question",
       "name": "Why does cloud platform engineering still matter?",
       "acceptedAnswer": {
         "@type": "Answer",
-        "text": "Cloud platform engineering provides the APIs, runtime reliability, observability, deployment automation, and secure operations needed to scale AI initiatives in production."
+        "text": "AI initiatives still depend on strong APIs, runtime reliability, observability, deployment automation, and secure operations. Without these foundations, most enterprise AI pilots fail to scale into production."
+      }
+    },
+    {
+      "@type": "Question",
+      "name": "What kinds of companies does Umplify help?",
+      "acceptedAnswer": {
+        "@type": "Answer",
+        "text": "Umplify works best with software-led businesses, startups, and enterprise teams in Toronto, the Greater Toronto Area, and across Canada that need AI transformation, application modernization, systems integration, or Azure platform engineering support."
+      }
+    },
+    {
+      "@type": "Question",
+      "name": "How does an Umplify engagement typically start?",
+      "acceptedAnswer": {
+        "@type": "Answer",
+        "text": "Most engagements begin with a focused discovery conversation followed by a short paid assessment (AI readiness, agentic workflow discovery, or platform architecture review). The output is a prioritized opportunity map, a delivery framing, and a clear recommendation for what to build first."
+      }
+    },
+    {
+      "@type": "Question",
+      "name": "Do you work with Microsoft Azure exclusively?",
+      "acceptedAnswer": {
+        "@type": "Answer",
+        "text": "Our deepest expertise is in Microsoft Azure, .NET, and the Azure AI stack, including Azure OpenAI, Container Apps, Functions, API Management, and Bicep or Terraform automation. We can integrate with non-Azure systems where the architecture requires it."
+      }
+    },
+    {
+      "@type": "Question",
+      "name": "How do you protect sensitive enterprise data when integrating AI?",
+      "acceptedAnswer": {
+        "@type": "Answer",
+        "text": "We design private knowledge access patterns, permission-aware retrieval, and controlled execution boundaries so AI can use internal context without compromising trust, security, or regulatory posture. Sensitive data stays inside your Azure tenant."
       }
     }
   ]