Current status of Jonathan Ross’ founded?
Executive summary
Groq, the AI chip company founded by Jonathan Ross in 2016, remains a standalone corporate entity on paper even as its core inference technology and key engineering team — including Ross — have moved into a high-value licensing arrangement with Nvidia announced in late December 2025; Groq framed the deal as a non‑exclusive license for its IP while confirming that Ross and senior staff will join Nvidia [1] [2] [3]. Reporting also describes the transaction as worth roughly $20 billion in cash for Groq’s assets, with some parts of the company such as a nascent cloud business explicitly excluded from the agreement [3] [4].
1. The origin story: Ross built Groq after leaving Google
Jonathan Ross co‑founded Groq in 2016 with Douglas Wightman after both worked on Google’s TPU project, positioning Groq as a startup focused on a single‑cycle, streaming inference ASIC originally called the Tensor Streaming Processor and later rebranded as the Language Processing Unit [1]. Public profiles and corporate bios uniformly describe Ross as Groq’s founder and long‑time CEO, and the company grew into a well‑funded, visible player in AI inference hardware by the mid‑2020s [5] [6] [7].
2. The December 2025 deal: licensing IP and personnel, not a conventional buyout
Multiple outlets report that Nvidia struck a deal in late December 2025 to license Groq’s inference technology for about $20 billion in cash, with the transaction structured as a non‑exclusive licensing agreement rather than a traditional acquisition of the corporate entity itself [3] [2] [8]. As part of that arrangement Groq’s founder Jonathan Ross, Groq president Sunny Madra and other senior engineers are set to join Nvidia — a talent and IP transfer that industry coverage treats as functionally closing Groq’s engineering chapter even as the company asserts independent status [9] [2].
3. What “independent” means here — ambiguity and exclusions
Groq and some reports stress that the company will “continue to operate as an independent company,” while other reporting clarifies that Nvidia will acquire “all of Groq’s assets” except for a recently launched cloud business that is explicitly excluded from the deal — an arrangement that blends IP licensing, staff transitions and retained corporate fragments rather than a clean takeover [1] [4]. That hybrid structure leaves open questions about Groq’s future operations, governance and product roadmap now that its founding CEO and core engineering team are moving to Nvidia [2] [3].
4. Industry reaction and internal consequences
Coverage flagged pushback inside Silicon Valley over the terms and human impact of the deal, with critics saying employees were “left out in the cold” even as founders and top engineers secured positions at Nvidia — a common critique when large players license technology and absorb talent without a full corporate merger that preserves broader employee equity or roles [9]. Financial and trade press framed the move as a continuation of big‑tech strategies to secure IP and people quickly, and analysts noted the deal further concentrates inference capabilities under Nvidia’s ecosystem [3] [2].
5. Limits of the public record and what cannot be asserted
Available reporting documents Ross’s role in the deal and the licensing terms at a high level, but public sources do not provide a definitive company‑by‑company chart of Groq’s remaining leadership, the fate of every business line, or the precise financial mechanics for employees and minority investors; those details remain opaque in the reporting cited and cannot be asserted here without further primary disclosures [2] [4]. Official statements and subsequent filings would be required to settle questions about Groq’s post‑deal board, operating autonomy, employee retention packages and the long‑term fate of Groq’s independent brand [1] [3].