{
  "id": 15373627,
  "title": "How Ankur Gupta is making AI model deployment more reliable",
  "url": "https://urgent.news/2026/10/18/how-ankur-gupta-is-making-ai-model-deployment-more-reliable",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-18T07:46:32.000Z",
  "source": {
    "name": "Jerusalem Post Tech & Start-Ups",
    "slug": "jerusalem-post-tech-start-ups",
    "url": "https://www.jpost.com/business-and-innovation/tech-and-start-ups/article-912017"
  },
  "original_language": "en",
  "account": null,
  "summary": "Ankur Gupta, a Senior Staff Technical Program Manager at LinkedIn, has made significant strides in making AI model deployment more reliable through his work on a control plane that coordinates the steps required to put an AI model into production. The control plane validates, tests, and manages a gradual rollout of the model, with systems responsible for computing capacity, networking, traffic routing, and other services all playing a part. Gupta organized the reliability effort around the full deployment experience, aligning engineering teams on a shared classification of failures, clear ownership across dependencies, and priorities for corrective work. By building dashboards that show where deployments fail and where time is being spent, the teams were able to make recurring patterns visible enough to prioritize, rather than requiring another investigation from scratch each time a customer encountered a problem. Trustworthy measurement also required separating different kinds of outcomes, with Gupta distinguishing improvements in classification from improvements in the platform itself. The result is a more dependable path to production, with clear responsibility for each failure domain and accountable owners for corrective actions.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}