vCluster Labs growth playbook

Add an "AI Cloud and AI Factory" column or tab: what is included (vMetal provisioning, managed Kubernetes, Slurm, Run:ai and inference Stacks, vBilling, 24/7 infra engineers), how it is priced (per GPU or node, in words if not numbers), the proof of value ("<45 days from decision to production") and the operators that run it (Nebius, Nscale). Give it its own call to action, "Plan your AI cloud launch", that routes to AI Infrastructure Engineers, and fix the "mission-critrical" typo while the page is open.

Buyers check pricing before they agree to a call, and the AI cloud deals are the large ones. A page that describes their purchase in their words turns ABM clicks into meetings instead of bounces, and gives the AEs' Business Value Assessment a public starting point.

[ ] Tab name: AI Cloud and AI Factory
[ ] Included: vMetal, managed K8s, Slurm, Run:ai and inference Stacks, vBilling
[ ] How it is priced (per GPU or node)
[ ] Proof of value: <45 days to production
[ ] Operators that run it
[ ] "Plan your AI cloud launch" routed to AI Infrastructure Engineers
[ ] Fix "mission-critrical"
[ ] HubSpot source tag for ABM traffic

• Publishing prices limits negotiation

  Describe the pricing model and a starting point, not a rate card.

• Confuses Kubernetes-only buyers

  Keep the current plans as the default tab for that audience.

• Your websitePricing page (checked October 23, 2026): titled "Flexible Plans for Kubernetes Multi-Tenancy" with the headline "Scale Kubernetes Without Scaling Costs"; three plans: Open Source, Free ($0, "Unlimited virtual clusters for up to 64 CPUs or 32 GPUs, and max 1 HA vCluster") and Enterprise ("Chat with Sales"). Outside the site-wide menu, the page never mentions vMetal, Slurm, Stacks, Run:ai, inference, NVIDIA or AI clouds; GPUs appear once, in the free-plan limit. The Enterprise card reads "mission-critrical".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source
• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source
• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source

Make the card a CMS item fed by the newest press release or changelog entry, with a working link, and add it to the launch checklist so every launch updates the menu, the changelog, LinkedIn, X and the newsletter on the same day.

The menu is seen on every page; a stale card makes a fast-shipping company look slow.

• Too many launches

  Show the latest major one.

• Your websiteThe Product menu's "LATEST RELEASE" card shows "vCluster v0.27 - Dedicated Clusters with Private Nodes" and its "Read more" links to "#". On GitHub the latest stable release is v0.37.4 (October 20, 2026) with v0.38 alphas this week; loft-sh/vcluster has 11,354 stars. vMetal and vNode are marketed on separate domains (vmetal.ai, vnode.com).Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• HiringSocial Media Manager (US, posted October 7, 2026): "the reason vCluster shows up in every conversation that moves the AI infrastructure industry, within hours instead of weeks"; follows "SemiAnalysis, Hacker News, X, funding announcements"; "The 24 to 48 Hour Window"; ghostwrites for the CEO and leaders; "Own the vCluster LinkedIn and X accounts, including a heavy launch and event calendar"; covers events like KubeCon live.Source

Sync careers pages to the Ashby board: remove or redirect closed roles to /careers with a "this role is filled" note, drop them from the sitemap, and add JobPosting structured data with valid-through dates to open ones.

Clean listings send candidates to the roles that matter now and keep the employer brand sharp during heavy hiring.

• Losing old links

  Redirect to /careers, not 404.

• Your websiteSitemap (1,168 URLs, checked October 23, 2026): 504 blog posts, 120 event pages, 66 /uses pages (including Rafay, Mirantis, OpenShift and VMware alternatives, and two near-identical GPU-as-a-service pages), 17 guides, 7 ebooks, 12 case studies, 214 author pages. Many older posts cover developer tooling (DevPod vs Gitpod, Docker Compose alternatives, Pulumi vs Terraform). 91 of the 124 careers URLs are for roles no longer on the Ashby board, and they still load (for example a Staff Product Manager post).Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source
• HiringCompany description in every post: "raised over $28M" including Khosla Ventures; customers turn raw GPUs into cluster products "they can sell in days instead of spending 12+ months building it themselves"; "over 100,000 GPUs and 1 million CPUs across 50+ AI clouds and Fortune 500 companies"; vCluster open source has "11,000+ GitHub stars and 40M+ tenant clusters created since 2021".Source

Check every profile name, handle, banner and bio against vCluster Labs, claim matching handles where possible, update the site links, and pin a current launch on each profile.

Consistent names make the brand easier to find, tag and follow, which matters most for the 24 to 48 hour posts.

• Lose followers on a rename

  Keep the account, change the display name and handle.

• Social profilesYouTube @vcluster: 2,860 subscribers, 324 videos, about 768,000 views. The site's social links still point to the LinkedIn page /company/loft-sh and to X as loft_sh; the newsletter form is HubSpot; the community Slack lives at slack.loft.sh.Source
• HiringSocial Media Manager (US, posted October 7, 2026): "the reason vCluster shows up in every conversation that moves the AI infrastructure industry, within hours instead of weeks"; follows "SemiAnalysis, Hacker News, X, funding announcements"; "The 24 to 48 Hour Window"; ghostwrites for the CEO and leaders; "Own the vCluster LinkedIn and X accounts, including a heavy launch and event calendar"; covers events like KubeCon live.Source

Build a 10-question self-assessment, one per ClusterMAX dimension, that gives an operator a gap report showing which vCluster, vNode or vMetal capability closes each gap, with a call to book a review with an AI Infrastructure Engineer. Gate the full report with an email.

Every AI cloud wants a higher rating; a personal gap report is a reason to talk that does not feel like a sales pitch.

• Seen as claiming to grade providers

  Call it a self-check against public criteria, not a rating.

• Your websiteClusterMAX guide: "ClusterMAX, published by SemiAnalysis, has become the de facto standard for evaluating GPU cloud infrastructure"; ten dimensions (Security, Lifecycle, Orchestration, Storage, Networking, Reliability, Monitoring, Pricing, Partnerships, Availability); the Security criterion lists "Strong isolation between tenants. i.e. use vCluster Private Nodes"; "Many providers plateau in their ratings ... because they lack a managed platform layer." The only action is "Get a Demo".Source
• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source
• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source

Send product usage for Free accounts into Salesforce, flag accounts above 75% of a limit or asking for a second HA vCluster, and have an AE offer help with a technical note rather than a pitch. Measure conversion from these product-qualified leads.

Usage is the most honest buying signal, and a helpful message at the limit matches the no-pressure promise.

• Users feel watched

  Say in the plan terms that usage informs support.

• Your websiteFree plan page: "A Free Plan That Is Actually Useful"; "Get your first vCluster in 5 minutes. No credit card. No forced demo. No sales pressure."; vCluster Cloud (hosted platform) or self-hosted. A "Show HN: vCluster Free" post reached Hacker News.Source
• Your websitePricing page (checked October 23, 2026): titled "Flexible Plans for Kubernetes Multi-Tenancy" with the headline "Scale Kubernetes Without Scaling Costs"; three plans: Open Source, Free ($0, "Unlimited virtual clusters for up to 64 CPUs or 32 GPUs, and max 1 HA vCluster") and Enterprise ("Chat with Sales"). Outside the site-wide menu, the page never mentions vMetal, Slurm, Stacks, Run:ai, inference, NVIDIA or AI clouds; GPUs appear once, in the free-plan limit. The Enterprise card reads "mission-critrical".Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source

Tier the rated providers by size and rating, build a one-page gap brief for each top-tier account from public criteria, and run LinkedIn ads, AE outreach and an invitation to an operator roundtable. Track pipeline per tier.

Precise accounts with a reason to care beat broad awareness, which is what the role asks for.

• Providers already on vCluster

  Exclude customers and route them to CSEs.

• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source
• Your websiteClusterMAX guide: "ClusterMAX, published by SemiAnalysis, has become the de facto standard for evaluating GPU cloud infrastructure"; ten dimensions (Security, Lifecycle, Orchestration, Storage, Networking, Reliability, Monitoring, Pricing, Partnerships, Availability); the Security criterion lists "Strong isolation between tenants. i.e. use vCluster Private Nodes"; "Many providers plateau in their ratings ... because they lack a managed platform layer." The only action is "Get a Demo".Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source

Run one monthly livestream with an operator customer and a vCluster engineer on one problem (tenant isolation, Slurm, inference per token). Give registrants a recording, a guide and an AE follow-up within two days.

Operators trust other operators, and a regular slot builds an audience a one-off webinar cannot.

• Low live attendance

  Clip and post the recording the same day.

• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• Your websiteSitemap (1,168 URLs, checked October 23, 2026): 504 blog posts, 120 event pages, 66 /uses pages (including Rafay, Mirantis, OpenShift and VMware alternatives, and two near-identical GPU-as-a-service pages), 17 guides, 7 ebooks, 12 case studies, 214 author pages. Many older posts cover developer tooling (DevPod vs Gitpod, Docker Compose alternatives, Pulumi vs Terraform). 91 of the 124 careers URLs are for roles no longer on the Ashby board, and they still load (for example a Staff Product Manager post).Source
• Social profilesYouTube @vcluster: 2,860 subscribers, 324 videos, about 768,000 views. The site's social links still point to the LinkedIn page /company/loft-sh and to X as loft_sh; the newsletter form is HubSpot; the community Slack lives at slack.loft.sh.Source

Publish a partner program page: who it is for, tiers and benefits, the registration path, certification, and a deal registration form tied to Salesforce. Add partner-ready assets (pitch, reference architectures, pricing guidance).

Resellers decide quickly whether a vendor is easy to sell; a clear program shortens recruiting and the first registered deal.

• Channel conflict

  Clear registration rules and comp-neutral co-sell.

• Your websiteEcosystem and Partners page: logos for marketplaces, system integrators and "Reseller Partners", each with "Visit site", and one action, "Contact Partner Group". No partner program, tiers, deal registration, partner portal or validated-stack pages.Source
• HiringPartner hiring: a Sr. Global Alliance Director, NVIDIA and AI Clouds (co-sell plays across NVIDIA Cloud Partners, sovereign AI programs and enterprises; "demand generation that NVIDIA sellers can run without us in the room"); a Sr. Director, VAR & Channel Partnerships for "a first-year partner program" ("Time from signature to first dollar is your headline metric"; deal registration, tiers, MDF); a Sr. Director, OEM, Storage and ISV; and a first Sr. Partner Solutions Engineer to turn partner deployments into "validated reference architectures" with "bills of material, deployment guides, demo environments, and benchmarks".Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source

Publish a "Validated stacks" section: one page per stack (server, storage, network, vCluster products), with a diagram, bill of materials, deployment guide and a benchmark. Co-brand with each partner and give their sellers the link.

Stack pages let partner sellers quote with confidence and rank for buyers searching for a specific hardware combination.

• Benchmarks age

  Date them and rerun on major releases.

• HiringPartner hiring: a Sr. Global Alliance Director, NVIDIA and AI Clouds (co-sell plays across NVIDIA Cloud Partners, sovereign AI programs and enterprises; "demand generation that NVIDIA sellers can run without us in the room"); a Sr. Director, VAR & Channel Partnerships for "a first-year partner program" ("Time from signature to first dollar is your headline metric"; deal registration, tiers, MDF); a Sr. Director, OEM, Storage and ISV; and a first Sr. Partner Solutions Engineer to turn partner deployments into "validated reference architectures" with "bills of material, deployment guides, demo environments, and benchmarks".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source
• Your websiteEcosystem and Partners page: logos for marketplaces, system integrators and "Reseller Partners", each with "Visit site", and one action, "Contact Partner Group". No partner program, tiers, deal registration, partner portal or validated-stack pages.Source

Package a kit for NVIDIA sellers: a one-page "when to bring vCluster" guide, a ten-minute demo video, a joint customer story, a first-meeting deck and a registration path that credits the NVIDIA seller. Train one region first.

Partners sell what is easy to sell; a self-contained kit multiplies the alliance team.

• Field ignores the kit

  Tie it to a joint incentive and a named champion.

• HiringPartner hiring: a Sr. Global Alliance Director, NVIDIA and AI Clouds (co-sell plays across NVIDIA Cloud Partners, sovereign AI programs and enterprises; "demand generation that NVIDIA sellers can run without us in the room"); a Sr. Director, VAR & Channel Partnerships for "a first-year partner program" ("Time from signature to first dollar is your headline metric"; deal registration, tiers, MDF); a Sr. Director, OEM, Storage and ISV; and a first Sr. Partner Solutions Engineer to turn partner deployments into "validated reference architectures" with "bills of material, deployment guides, demo environments, and benchmarks".Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source

Publish the plan week by week: prerequisites (nodes, network, storage), what vCluster does, what the customer does, the go-live checklist and the exit criteria. Offer it as the next step on pricing and AI cloud pages.

A visible plan lowers perceived risk for an infrastructure decision and qualifies buyers who are ready.

• Plans vary

  Show a standard plan and note what extends it.

• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source
• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source

Add expansion triggers to every success plan (new GPU capacity, a new region, a new managed service, Free-to-Enterprise users in the same company) and review them monthly with the AE. Offer Stacks as the path to new services.

Expansion from happy customers is the cheapest revenue there is, and triggers make it routine.

• Feels pushy

  Lead with the customer's own roadmap.

• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source

Tag the developer posts, give them a call to action for vind and the Free plan instead of a demo, and keep demo calls to action on AI infrastructure content. Report the two funnels separately.

Each audience gets a next step it will take, and demo numbers stop being diluted by developer traffic.

• Old traffic declines

  Refresh rather than delete.

• Your websiteSitemap (1,168 URLs, checked October 23, 2026): 504 blog posts, 120 event pages, 66 /uses pages (including Rafay, Mirantis, OpenShift and VMware alternatives, and two near-identical GPU-as-a-service pages), 17 guides, 7 ebooks, 12 case studies, 214 author pages. Many older posts cover developer tooling (DevPod vs Gitpod, Docker Compose alternatives, Pulumi vs Terraform). 91 of the 124 careers URLs are for roles no longer on the Ashby board, and they still load (for example a Staff Product Manager post).Source
• Your websiteFree plan page: "A Free Plan That Is Actually Useful"; "Get your first vCluster in 5 minutes. No credit card. No forced demo. No sales pressure."; vCluster Cloud (hosted platform) or self-hosted. A "Show HN: vCluster Free" post reached Hacker News.Source
• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source

Write four new stories (a telco, a neocloud, an inference partner and an enterprise AI factory), each with one hard number (time to first tenant, GPU utilization, revenue launched), and add them to the pricing and solution pages.

Numbers make proof portable, and a story per segment covers each AE conversation.

• Approval delays

  Offer anonymised versions.

• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source
• Your websiteSitemap (1,168 URLs, checked October 23, 2026): 504 blog posts, 120 event pages, 66 /uses pages (including Rafay, Mirantis, OpenShift and VMware alternatives, and two near-identical GPU-as-a-service pages), 17 guides, 7 ebooks, 12 case studies, 214 author pages. Many older posts cover developer tooling (DevPod vs Gitpod, Docker Compose alternatives, Pulumi vs Terraform). 91 of the 124 careers URLs are for roles no longer on the Ashby board, and they still load (for example a Staff Product Manager post).Source
• HiringSales hiring: five Enterprise Account Executive posts (October 6-20, 2026) who build a "Business Value Assessment"; a Regional Sales Director, West to "Operationalize 'Land & Expand'" and "capitalize on our massive NRR potential"; a Revenue Operations Specialist (Deal Desk & Reporting) owning quote-to-cash and "the Stripe to Salesforce ARR sync".Source

Keep a running list of vCluster's positions (isolation, utilization, sovereign clouds, per-token revenue), draft two angles within a day of each news item, and post from the CEO's account first. Clip talks into short video the same day.

Founder posts reach operators who ignore company pages, and fast takes make vCluster part of the conversation.

• Hot takes misfire

  Quick gut check before sensitive posts.

• HiringSocial Media Manager (US, posted October 7, 2026): "the reason vCluster shows up in every conversation that moves the AI infrastructure industry, within hours instead of weeks"; follows "SemiAnalysis, Hacker News, X, funding announcements"; "The 24 to 48 Hour Window"; ghostwrites for the CEO and leaders; "Own the vCluster LinkedIn and X accounts, including a heavy launch and event calendar"; covers events like KubeCon live.Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• Your websiteClusterMAX guide: "ClusterMAX, published by SemiAnalysis, has become the de facto standard for evaluating GPU cloud infrastructure"; ten dimensions (Security, Lifecycle, Orchestration, Storage, Networking, Reliability, Monitoring, Pricing, Partnerships, Availability); the Security criterion lists "Strong isolation between tenants. i.e. use vCluster Private Nodes"; "Many providers plateau in their ratings ... because they lack a managed platform layer." The only action is "Get a Demo".Source

Give each product a section on vcluster.com (or clear, consistent cross-links), add them to pricing and the AI cloud tab, and keep one demo path and one analytics view.

One site makes the platform story clear and keeps all the links and authority in one place.

• Losing rankings on redirect

  Map page by page.

• Your websiteThe Product menu's "LATEST RELEASE" card shows "vCluster v0.27 - Dedicated Clusters with Private Nodes" and its "Read more" links to "#". On GitHub the latest stable release is v0.37.4 (October 20, 2026) with v0.38 alphas this week; loft-sh/vcluster has 11,354 stars. vMetal and vNode are marketed on separate domains (vmetal.ai, vnode.com).Source
• Your websitePricing page (checked October 23, 2026): titled "Flexible Plans for Kubernetes Multi-Tenancy" with the headline "Scale Kubernetes Without Scaling Costs"; three plans: Open Source, Free ($0, "Unlimited virtual clusters for up to 64 CPUs or 32 GPUs, and max 1 HA vCluster") and Enterprise ("Chat with Sales"). Outside the site-wide menu, the page never mentions vMetal, Slurm, Stacks, Run:ai, inference, NVIDIA or AI clouds; GPUs appear once, in the free-plan limit. The Enterprise card reads "mission-critrical".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source

Publish a Stacks catalog with partner listings (inference, Run:ai, Slurm, notebooks), each with a one-click deploy, pricing guidance for the operator and a partner page. Promote it as the fastest way to add a managed service.

A catalog turns partners into a growth channel and makes vCluster harder to replace.

• Low-quality listings

  Validate every Stack before listing.

• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source
• HiringPartner hiring: a Sr. Global Alliance Director, NVIDIA and AI Clouds (co-sell plays across NVIDIA Cloud Partners, sovereign AI programs and enterprises; "demand generation that NVIDIA sellers can run without us in the room"); a Sr. Director, VAR & Channel Partnerships for "a first-year partner program" ("Time from signature to first dollar is your headline metric"; deal registration, tiers, MDF); a Sr. Director, OEM, Storage and ISV; and a first Sr. Partner Solutions Engineer to turn partner deployments into "validated reference architectures" with "bills of material, deployment guides, demo environments, and benchmarks".Source
• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source

Publish a yearly benchmark: median time to tenant, utilization by tenancy model, mix of Kubernetes, Slurm and inference, and what top operators do differently. Launch it with a livestream, press and the SemiAnalysis community.

Original data earns press and links and gives AEs a reason to call every operator.

• Customer data concerns

  Opt-in only and aggregate.

• HiringCompany description in every post: "raised over $28M" including Khosla Ventures; customers turn raw GPUs into cluster products "they can sell in days instead of spending 12+ months building it themselves"; "over 100,000 GPUs and 1 million CPUs across 50+ AI clouds and Fortune 500 companies"; vCluster open source has "11,000+ GitHub stars and 40M+ tenant clusters created since 2021".Source
• Your websiteClusterMAX guide: "ClusterMAX, published by SemiAnalysis, has become the de facto standard for evaluating GPU cloud infrastructure"; ten dimensions (Security, Lifecycle, Orchestration, Storage, Networking, Reliability, Monitoring, Pricing, Partnerships, Availability); the Security criterion lists "Strong isolation between tenants. i.e. use vCluster Private Nodes"; "Many providers plateau in their ratings ... because they lack a managed platform layer." The only action is "Get a Demo".Source
• HiringSr. Demand Generation Manager (US, posted October 7, 2026): "building the targeted account motion that turns the right platform engineers and infrastructure leaders at the right enterprises and AI Cloud providers into pipeline. There's no existing playbook." Owns LinkedIn, paid search, content syndication, "the livestream and webinar motion" ("the follow-through that turns registrants into pipeline"), best-fit customer profile and target accounts with sales, and "pipeline influence, new opportunities sourced, channel efficiency". Stack: HubSpot or Marketo, Salesforce, intent data (6sense, Bombora), an ABM platform.Source

Run a shared demo environment with a GPU partner: a prospect gets a guided, time-limited tenant cluster with a Stack deployed, sees the operator view, and books a POV from inside it.

Seeing the product on real GPUs shortens evaluation for the biggest deals.

• GPU cost

  Time limits and best-fit customer profile gating.

• Your websiteFree plan page: "A Free Plan That Is Actually Useful"; "Get your first vCluster in 5 minutes. No credit card. No forced demo. No sales pressure."; vCluster Cloud (hosted platform) or self-hosted. A "Show HN: vCluster Free" post reached Hacker News.Source
• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source
• Your websitePress room, October 2026: Stacks generally available in vCluster Platform v4.12 (October 8); managed NVIDIA Run:ai clusters on the newly open-sourced Run:ai (October 13); Saturn Cloud per-token inference on vCluster Stacks (October 14); LTS releases (October 18); Deutsche Telekom's Industrial AI Cloud automated by vCluster with NVIDIA DSX OS (October 21).Source

Release a small open-source CLI that runs read-only checks on a cluster (shared nodes, network policies, API access across tenants) and prints a report, with links to how vCluster solves each gap.

Open-source tools spread through engineers and put vCluster in front of every operator who runs them.

• Seen as fear marketing

  Keep checks factual and open.

• HiringCompany description in every post: "raised over $28M" including Khosla Ventures; customers turn raw GPUs into cluster products "they can sell in days instead of spending 12+ months building it themselves"; "over 100,000 GPUs and 1 million CPUs across 50+ AI clouds and Fortune 500 companies"; vCluster open source has "11,000+ GitHub stars and 40M+ tenant clusters created since 2021".Source
• Your websiteClusterMAX guide: "ClusterMAX, published by SemiAnalysis, has become the de facto standard for evaluating GPU cloud infrastructure"; ten dimensions (Security, Lifecycle, Orchestration, Storage, Networking, Reliability, Monitoring, Pricing, Partnerships, Availability); the Security criterion lists "Strong isolation between tenants. i.e. use vCluster Private Nodes"; "Many providers plateau in their ratings ... because they lack a managed platform layer." The only action is "Get a Demo".Source
• Your websiteThe Product menu's "LATEST RELEASE" card shows "vCluster v0.27 - Dedicated Clusters with Private Nodes" and its "Read more" links to "#". On GitHub the latest stable release is v0.37.4 (October 20, 2026) with v0.38 alphas this week; loft-sh/vcluster has 11,354 stars. vMetal and vNode are marketed on separate domains (vmetal.ai, vnode.com).Source

Turn the internal notes into a public field guide (one chapter per layer), keep it in the docs, and have each engineer add a chapter per deployment. Use it in recruiting too.

Hands-on guides rank for hard technical searches and show the depth behind "40+ hardcore infra engineers".

• Engineers short on time

  One chapter a month, edited by marketing.

• HiringCustomer engineering: AI Infrastructure Engineers who work "pre-sale as part of a proof of value engagement scoped to reach production", from bare metal GPU nodes (GPU Operator, RDMA/InfiniBand) to a validated environment ("the playbooks you develop will scale the motion"); Customer Success Engineers who own onboarding and success plans and spot "expansion opportunities" and churn. A Senior Product Manager will "Sharpen the narrative for technical buyers".Source
• HiringCompany description in every post: "raised over $28M" including Khosla Ventures; customers turn raw GPUs into cluster products "they can sell in days instead of spending 12+ months building it themselves"; "over 100,000 GPUs and 1 million CPUs across 50+ AI clouds and Fortune 500 companies"; vCluster open source has "11,000+ GitHub stars and 40M+ tenant clusters created since 2021".Source
• Your websiteHomepage: "Run your AI cloud like a hyperscaler. One platform to turn raw GPUs into every kind of cluster, from bare metal up to Kubernetes, Slurm, inference, and more"; a banner for vMetal; proof points "1 min To spin up isolated K8s environments", "<45 Days from decision to production launch", "31% Of Nebius Token Factory running on vCluster"; "Named in the ClusterMAX security criteria"; talks with NVIDIA, Adobe and Nscale.Source