Leadership1 distinct publisher3 min readUpdated
A program manager cleared the H-1B lottery in March and called it a weight off his mind. He still will not buy a house or start a family, and he is still pricing Canada.
The Board Room · Leadership desk
Compiled by The Board RoomSomething wrong?How this is made
Madhur Mehta, a 35-year-old Indian national working as a program manager at Amazon in Seattle, told Business Insider that being selected in the H-1B lottery in March felt like someone lifting a 100-pound weight off his mind [1][7]. In the same essay he said he and his fiancee explored buying a house last year and could not commit, and that starting a family waits on a different visa entirely [12][14]. That gap between a sponsored, retained employee and a committed one is where attrition actually begins, and it does not appear in any HR dashboard.
The sequence matters for anyone who thinks the sponsorship line item closes the issue. Mehta arrived in August 2021 for a master's degree, graduated from UT Dallas in 2023 in a STEM field that gave him three years of Optional Practical Training, and said he struggled to find an employer willing to sponsor him beyond that window until he joined Amazon in 2025 [2][3][4]. That is roughly two years of his work authorization spent hunting for a sponsor rather than compounding inside a company [20]. While Amazon was filing, he learned the government was changing the lottery to sort registrants into four wage-level categories, with higher categories getting better odds, so he built a Plan B and a Plan C: an EB-1A extraordinary ability petition, and jobs in Canada and Australia [5][6].
He won, and he is still working the Plan B. Mehta said he intends to pursue the EB-1A and is glad the H-1B buys him time to build the profile for it [8]. Read that as an employee whose stated three-year horizon is a bridge, not a destination.
The household arithmetic is the load-bearing part. Lose the job on an H-1B and there are generally 60 days to find another [9]. Mehta got engaged in July 2025 to a fellow Indian national who has one remaining attempt in the lottery before her work authorization expires; if she is not selected she may become his dependent on an H-4, which does not automatically carry work authorization, and enrolling in another program is complicated by recent student visa changes [10][11]. So a single layoff in a two-immigrant household can put the family's legal residence and both careers on a two-month clock [21]. No retention bonus prices that risk out.
What the employer gets in the meantime is real but narrow. Mehta said the US has been worth it, that the network, exposure and credibility of working there compound, and that Amazon's offices in India would not be the same [15]. What it does not get is duration. He said friends of more than ten years are still renting because of uncertainty, and that Canada and Europe, where the pathways seem more stable, come up in conversations with his fiancee [13][16]. The only thing holding him is switching cost: the difficulty of packing a life into two or three suitcases [17].
Watch three inputs. His fiancee's final lottery attempt, because a household in limbo makes relocation cheaper than staying [10]. The wage-level selection rules, which change who inside your org has odds [5]. And the student visa changes narrowing the fallback that used to absorb people between statuses [11]. Managers reading engagement scores should note that Mehta describes five years in survival mode and no rest, and calls the public version of his life half the story [18].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
The Business Insider essay is an as-told-to based on a conversation with Madhur Mehta, 35, an Indian national living in Seattle and working at Amazon as a program manager.
While Amazon was trying to get him an H-1B, he learned the government was changing the lottery system for H-1B registrations so candidates would be placed into four wage-level categories, with those in higher categories having a greater chance of selection.
He still plans to pursue the EB-1A and says he is glad the H-1B gives him more time to build up his profile for the application.
He says that if he gets his EB-1A he will be able to think about starting a family and buying a house, travel back to India without worrying about stamping, and worry less about what happens if he loses his job.
Mehta came to the US in August 2021 for his master's degree, with the goal of a stable career at a good company and supporting family in India.
He graduated with a master's in science and business analytics from UT Dallas in 2023; because the course was STEM, he was eligible for three years of student work authorization through Optional Practical Training.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single verified first-person account, no second source
All claims trace to one Business Insider as-told-to essay. The publisher verified only that the subject is employed by Amazon; dates, the lottery rule change, his fiancée's status, and the 60-day window are self-reported and edited for length. No regulatory document, legal counsel, employer statement, or second worker corroborates any element, and Amazon did not respond, so the record is internally consistent but thin and one-sided.
No measurable adoption signal
The cluster supplies no counts, rates, or disclosures that would let adoption be measured: no data on how many sponsored workers defer home purchases or relocate, no employer sponsorship volumes, and no selection statistics under the described wage-level lottery. The subject's reference to friends who still rent after ten years is anecdotal and unquantified, and inferring a population-level pattern from it would be guessing.
Systemic framing runs slightly ahead of one anecdote
The subject's own claims are modest, hedged, and mostly first-person, which keeps the gap small. The overstatement is structural rather than rhetorical: a single household's deferred house purchase and one uncorroborated account of a lottery rule change are being carried as a general proposition that sponsorship does not retain talent, without any population data or employer response to size the effect.
Moderate subject-side narrative incentive, employer absent
The subject is an identified individual with an interest in how his situation reads: he states he is actively building a profile for an EB-1A petition, and the essay format lets him present his own account without challenge. Business Insider's as-told-to genre rewards personal hardship narratives. Offsetting factors keep this mid-range: he does not solicit anything, credits the US tech premium, and the publisher disclosed both its verification step and Amazon's non-response rather than concealing the one-sidedness.
Low-moderate: coherent account, unverifiable specifics
Confidence is limited by one publisher, one subject, an unverified policy premise, and no adoption measurement. What raises it above the floor is that the essay's mechanics are specific and mutually consistent, the publisher verified employment and disclosed the employer's non-response, and the story's core assertions are framed as personal experience rather than as data.
leadership
Disney swaps raises for discounted stock and a full health-plan re-enrollment1 distinct publisher
leadership
Plan for both: AI reshapes the economy and most of this wave's capital is lost1 distinct publisher
leadership
31 years at Microsoft, then a layoff: what tech workers' regrets say about retention1 distinct publisher
leadership
The retention lever managers assumed they had: visa-dependent engineers are now leaving Big Tech first1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.