Key Highlights
- Gadi Hutt, who served as director of product and customer engineering at Annapurna Labs, has exited Amazon
- Hutt played a central role in developing Amazon’s Trainium AI chip platform
- This marks the second high-level departure from Annapurna in a seven-month period
- Former executive Rami Sinno joined Arm (ARM) in August 2025; AGI division leader Rohit Prasad exited late 2025
- AMZN shares finished Thursday’s session down 2%; up 3% year-over-year
Amazon (AMZN) shares retreated 2% during Thursday’s trading session.
According to reporting by The Information, Gadi Hutt has departed from his position as director of product and customer engineering at Annapurna Labs. Hutt served as a prominent figure in the development and promotion of Amazon’s Trainium AI processor family.
Back in 2015, Amazon purchased Annapurna Labs—an Israeli semiconductor company—for $350 million. Since then, the division has evolved into a critical component of Amazon’s strategy to design proprietary chips and reduce dependence on external semiconductor vendors.
This exit represents the second departure of a senior leader from Annapurna within seven months. Rami Sinno resigned in August 2025 and subsequently joined Arm Holdings (ARM).
Losing two critical chip division leaders within such a compressed timeframe raises significant questions about internal dynamics.
The talent drain extends beyond the semiconductor division. Rohit Prasad, who held the position of senior vice president and head scientist for Amazon’s artificial general intelligence efforts, also departed at the conclusion of 2025.
Multiple Levels of Leadership Turnover
The wave of departures spans various tiers of Amazon’s artificial intelligence infrastructure—ranging from Annapurna’s hardware development group to Prasad’s AGI research organization.
Amazon has remained silent on these exits, offering no official statements, and the motivations behind each departure remain undisclosed.
Trainium represents Amazon’s proprietary processor engineered specifically for AI model training tasks, serving as a cornerstone of the company’s mission to establish cloud computing infrastructure that isn’t wholly reliant on Nvidia’s technology.
Given his leadership position in product and customer engineering, Hutt maintained direct oversight of Trainium chip development processes and commercialization strategies.
Intense Competition for Specialized Expertise
The battle for AI semiconductor talent has intensified dramatically throughout the technology sector, with large technology corporations, emerging startups, and established chipmakers all vying for a limited pool of specialized engineers and product executives.
Sinno’s transition to Arm illustrates how rapidly elite professionals can migrate between competing organizations in this domain.
Amazon’s Annapurna Labs division has become essential to AWS operations, as the cloud platform leverages proprietary chips to deliver distinctive offerings to business clients.
Trainium processors have been marketed as economically viable alternatives to Nvidia’s GPU lineup for artificial intelligence training applications.
AMZN stock has gained 3% over the trailing twelve-month period, notwithstanding Thursday’s 2% decline triggered by the departure announcement.
