The Hidden Infrastructure of AI: A 20-Year Engineer’s EB1A After RFE
At a Glance: A Software Engineer from India, based in California, with 20 years of experience in Enterprise Distributed Systems with a focus on ML and AI Infrastructure, received EB1A approval after a Request for Evidence (RFE) on June 30, 2026. No patents were filed or granted.
Everyone talks about AI models. Few talk about the infrastructure that makes them run.
This engineer has spent twenty years building that infrastructure: the distributed systems that train, serve and scale AI at one of the world’s largest technology companies. The title says “Software Engineer.” The work says much more. The RFE gave the team the chance to prove it.
built systems that train and serve AI models at massive scale.
designed reliable systems spread across many machines.
improved how compute resources are used for AI workloads.
strengthens the infrastructure behind U.S. AI leadership.
We made the invisible infrastructure visible, showing how the applicant’s systems enable AI used by millions. The RFE response used independent expert letters and strategic documentation.
Every step followed our RFE playbook.
advancing the infrastructure that powers AI in the United States.
On June 30, 2026, the petition was approved.
22
Days
Every case is different. Past results do not guarantee a similar outcome.
Related Approval Stories: EB1A Approved After RFE: GenAI Engineering Manager · EB1A After RFE: AI Platform and Cloud Engineer · All EB1A approval stories.
Not sure if your profile qualifies? Start with a free EB1A eligibility check and see which EB1A criteria your evidence already meets.
Can AI infrastructure engineers qualify for EB1A?
Yes. Infrastructure is critical to AI progress.
Does a junior-sounding title hurt my EB1A?
Not if your evidence shows extraordinary work.
From evidence consolidation to national-interest framing, every detail is built for credibility and speed.