Will Marvell report any revenue contribution from Celestial AI products by Q4 FY2027?

The question asks for ANY revenue contribution, not meaningful revenue. Management's own timeline states 'meaningful revenue not expected until H2 FY2028' but the $500M ARR target by Q4 FY2028 requires a revenue ramp that likely starts in FY2027. To reach $500M ARR 12 months after the resolution date, there almost certainly needs to be initial shipments or sample revenue within FY2027. The AWS warrant extension to include Celestial AI's product swim lane confirms a specific customer with commercial intent. However, the technology is pre-production — a 6.4T engine was 'demonstrated' but not commercially shipped. The 12-month window from acquisition close to resolution deadline is tight for a pre-revenue technology company to generate even initial product revenue. Base rate for pre-revenue deep-tech acquisitions generating any revenue within 12 months is perhaps 40-50%, but Marvell's integration track record and existing customer relationships (AWS) pull this higher.

Management explicitly stated 'meaningful revenue not expected until H2 FY2028' — this is the clearest signal on timeline. While the question only requires ANY revenue (including sample shipments), the distinction between 'meaningful' and 'any' may be smaller than it appears for a photonic fabric product. This is silicon/photonics hardware — not software that can soft-launch. It requires production qualification, supply chain ramp, customer validation testing, and integration into data center architectures. Even 'sample revenue' requires production-grade parts. The Inphi comparison is instructive but misleading — Inphi was already generating hundreds of millions in revenue at acquisition. Celestial AI has $0 revenue and an unproven technology at commercial scale. The 380-person team and $50M OpEx suggests R&D-heavy, not commercialization-ready. The earnout structure ($2.25B tied to milestones) may actually indicate Marvell's own uncertainty about timeline. I weight management's explicit 'H2 FY2028' timeline heavily — companies rarely outperform their own stated timelines for pre-revenue technology.

This is a genuinely uncertain question where the key variable — semiconductor product commercialization timeline — has wide variance. The resolution criteria explicitly include 'initial customer shipments or sample revenue' which meaningfully lowers the bar versus requiring production-scale revenue. The dossier notes a '6.4T engine demonstrated' which suggests technology readiness is beyond pure R&D. AWS's warrant extension creates both a willing early customer and commercial pressure to deliver. Marvell's existing sales infrastructure and customer relationships could accelerate time-to-first-revenue compared to a standalone startup. However, the Consolidation Calibrator's finding that this is 'qualitatively different' from all prior deals is important — the integration playbook may not apply to pre-revenue photonic technology. The committee's unresolved debate (visionary bet vs. reckless departure) reflects genuine uncertainty. I anchor on slightly above 50% because the question's low bar (any revenue) and Marvell's customer access provide a slight edge, but confidence is LOW given the technology uncertainty.

Management gave a clear timeline: meaningful revenue H2 FY2028. The $500M ARR target by Q4 FY2028 does imply earlier ramp, but 'ramp' likely means H1 FY2028 initial revenue, not FY2027. The acquisition closed Feb 2, 2026 — meaning FY2027 Q1 starts around May 2026 and Q4 ends January 2027. That is 8-12 months of integration before the resolution deadline. For a pre-revenue photonic interconnect technology, going from $0 to any revenue in under a year post-close is aggressive. Semiconductor qualification cycles alone typically take 6-12 months. The AWS commercial interest is real but a warrant extension is not a purchase order. Marvell's track record with Inphi is irrelevant — Inphi already had products and customers. The Myth Meter's identification of 'home run' framing as narrative overreach is apt — management enthusiasm does not compress physics and qualification timelines.

The key nuance is what counts as 'revenue contribution.' The resolution criteria say 'including initial customer shipments or sample revenue, but excluding NRE or internal development.' Sample revenue from early-access or evaluation units to a customer like AWS could plausibly occur within FY2027. Marvell could ship evaluation units or small-volume initial production to key customers as part of qualification programs, and if those are revenue-recognized rather than treated as NRE, they would qualify. The AWS warrant extension specifically for Celestial AI products suggests active commercial engagement, not just theoretical interest. Companies often recognize small amounts of product revenue from sample shipments during qualification. However, the technology is novel (photonic fabric) and may require longer customer qualification. I lean slightly above 50% because the bar is genuinely low and Marvell has a motivated anchor customer, but the technology risk keeps this close to a coin flip.

I focus on the concrete facts: $0 revenue at close, 'meaningful revenue not until H2 FY2028', and the technology is 'commercially unproven.' The FY2028 growth framework attributes ~5-10pp of growth to Celestial AI, which on FY2027 revenue would imply $400-800M contribution in FY2028 — consistent with the $500M ARR target. But this is FY2028 revenue, not FY2027. The earnout structure with milestones through FY2029 suggests even Marvell expects a multi-year ramp. The fact that Q4 FY2026 earnings (March 2026) 'may provide first post-close update' means as of our analysis date, there is zero visibility into integration progress. The 113-day data staleness is material here. Without any post-close confirmation of progress, I weight the explicit management timeline and the inherent difficulty of commercializing novel photonics technology. Below 50%.

Pre-revenue technology, management says meaningful revenue H2 FY2028, but question asks for ANY revenue. AWS warrant extension suggests commercial engagement. The 12-month window from close to resolution is tight but not impossible for sample shipments. Slightly below 50% given hardware qualification timelines.

Photonic fabric is unproven commercially. Management explicitly guides meaningful revenue to H2 FY2028. Semiconductor products require qualification cycles of 6-12 months minimum. Even with Marvell's resources, going from $0 to any product revenue in ~12 months post-close is a stretch. The technology demonstration (6.4T engine) is not the same as production-ready product. Below 50%.

The $500M ARR by Q4 FY2028 target mathematically requires ramp to begin somewhere in FY2027, but 'beginning ramp' may mean design wins and initial orders, not recognized revenue. AWS interest is a positive signal but the bar to go from demonstrated technology to revenue-recognized shipments within 12 months of closing is high for novel hardware. Lean slightly below 50% but acknowledge significant uncertainty.