{"resourceId":"nps-secure-software-supply-chain-hackathon-dell-gb10-local-ai-20261104","versions":[{"version":"external-c9fa99c2cf529070adb213e445e1f43eb243accdc8f0306b362896a5de55882c","resource":{"id":"nps-secure-software-supply-chain-hackathon-dell-gb10-local-ai-20261104","title":"Naval Postgraduate School hackathon on software supply-chain risk used Dell Pro Precision GB10 deskside systems with locally hosted AI models","organization":"Naval Postgraduate School and NPS Foundation, with Dell Technologies, NVIDIA and Sterling","sector":"Graduate education and research (federal)","geography":"Monterey, California, United States","publishedAt":"2026-11-04","publicationDate":"2026-11-04","eventDate":null,"sourceName":"Dell Technologies blog (John Garrett)","sourceLabel":"Vendor blog","sourceUrl":"https://www.dell.com/en-us/blog/dell-and-nps-explore-emerging-technologies-during-campus-hackathon","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["data-security","accessibility-workforce"],"finding":"A Nov. 4, 2026 Dell blog says the NPS Foundation, with Dell, NVIDIA and Sterling, ran a Secure Software Supply Chain Hackathon at the Naval Postgraduate School asking how public or commercial information can assess the risk of malicious contributors in a software supply chain. Students used Dell Pro Precision with GB10 deskside devices and locally hosted AI models and tools, followed by a Dell technology tour. The post reports no solutions, metrics or evaluation of the approaches.","sledRelevance":"Limited: NPS is a federal graduate school, not SLED. Relevant to research universities as an example of deskside, locally hosted AI for sensitive security research and student hackathons.","evidence":"Vendor blog describing a one-day event; no measured outcomes or independent account.","architectureImplications":"Deskside GB10 systems allow local model hosting without cloud data egress.","governanceImplications":"Local hosting can keep sensitive research data on premises; governance details not given.","securityPrivacyImplications":"Focus on supply-chain contributor risk; event results not disclosed.","caveats":"Vendor marketing; federal not SLED; single event; no results.","streamIds":["dell","research"],"roles":{"sales":"Sales framing: this is a vendor-described hackathon at a federal graduate school, useful mainly as a talking point for research universities and cybersecurity programs considering deskside AI systems for teaching and sensitive projects. Dell supplied Pro Precision GB10 devices running local models. Do not claim the event produced results or that NPS adopted the systems; the blog reports neither. Use it to open discussion about local AI for classes and secure research labs.","engineering":"Engineering framing: students ran locally hosted AI models on Dell Pro Precision GB10 deskside systems to analyze software supply-chain contributor risk. Engineers should evaluate which model sizes fit the device memory, how datasets are staged locally, and how results are reproduced. Deskside hosting avoids cloud data movement but limits scale. Validate with a representative workload, such as repository metadata analysis, before recommending it for a lab.","delivery":"Delivery framing: running a hackathon or teaching lab on deskside AI systems requires imaging devices with approved models and tools, managing licenses, and resetting machines between events. Delivery teams would define who maintains the model catalog, how student data and outputs are retained or wiped, and how events are evaluated. The blog gives no outcomes, so plan measurable goals rather than relying on this account."},"roleDetails":{"sales":{"customerProblem":"Universities want hands-on AI and security learning environments that keep sensitive data local.","stakeholders":["Cybersecurity program director","Research computing","Department IT","Foundation or industry relations"],"discoveryQuestions":["Do courses need local model hosting?","What data cannot leave campus?","Who manages lab device images?"],"valueHypothesis":"Deskside systems may enable local AI exercises without cloud costs or data egress; not demonstrated here.","potentialEngagement":"A bounded pilot design for a deskside AI teaching or hackathon lab.","unsupportedClaims":["That the hackathon produced effective methods","Adoption by NPS","Cost advantages"]},"engineering":{"fit":"Fits small research groups, classes and events needing local model hosting.","architectureIntegration":"Standalone deskside devices with local models; integration with campus identity and storage unstated.","prerequisites":["Approved model catalog","Device imaging process","Dataset staging"],"constraints":["Device memory limits model size","Federal versus SLED applicability"],"security":"Local hosting limits data exposure; device wiping and access control still needed.","proofOfValue":["Run a representative supply-chain analysis workload on one device","Time setup and reset between sessions"]},"delivery":{"work":"Prepare, image and reset deskside AI devices for events and courses.","dependencies":["Model licenses","Campus IT security","Faculty organizers"],"ownership":"Department IT with faculty event organizers (interpretation).","skillsAdoption":"Student onboarding to local model tooling.","governanceCheckpoints":["Model and tool approval","Data retention and wipe policy"],"acceptanceCriteria":["Proposed: devices reset within a defined window after each event","Proposed: participant outcome survey"],"risks":["Vendor framing overstating impact","Unmanaged data left on devices"]}},"retrievedAt":"2026-11-05T04:05:14Z","enrichedAt":"2026-11-05T04:06:32Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Hands-on AI and security experience for graduate students.","procurementImplications":"Deskside AI systems are an alternative to shared clusters for small teaching labs; pricing not given.","operatingModelImplications":"Industry-supported hackathons as a vehicle for applied AI education.","sourceVerification":{"openedUrl":"https://www.dell.com/en-us/blog/dell-and-nps-explore-emerging-technologies-during-campus-hackathon","referenceExcerpt":"Participants worked with a Dell Pro Precision with GB10, deskside devices and locally hosted AI models and tools.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}