We make the tools that build the chip, not the chip - deposition, laser annealing, ion-beam coating, the epitaxy that grows the laser. Last year that was $664 million of revenue, down about 7%, with the semiconductor half nearly flat at $477 million while compound semi and data storage fell harder and pulled the total down. Let me give you some color on the part people ask about: this year multiple customers handed us more than $250 million of orders for indium phosphide laser tools, delivering from 2026 and ramping hard in 2027. Where do we actually sit? Strong incumbent in laser-facet coating, a second source in the epitaxy step behind an established competitor, strong in wet processing. We build to order, not to forecast - so the backlog is real, the lead times are long, and one customer keeps getting bigger: under 10% two years ago, 17% last year. That's the demand floor and the risk in the same sentence.
Let me give you some color on the concentration line. Our largest customer keeps growing as a share of revenue—under 10% two years ago, 11% in 2024, and 17% in 2025, and no other customer is above 10%. At the same time, China revenue fell about $80 million, or roughly 31%, in 2025 to about a quarter of total, which we tie to U.S. export controls. Because about 99% of our manufacturing is U.S.-based, we sit on the export‑control side of the split.
Let me give you some color: the more than $250 million of orders spans all three lines, but the strength is uneven, and we say so. On the epitaxy step a competitor holds a good incumbent position and we are the second source; on wet‑processing we hold a strong position with several leaders; on ion‑beam facet coating we are a strong incumbent at a number of key customers.
Give you some color on the more than $250 million of orders for indium phosphide laser tools — there are really three pieces: the MOCVD epitaxy platform that grows the active layer, the wet-processing and etch systems, and the ion-beam deposition tool for the laser facet coating. Deliveries start in 2026 and accelerate hard in 2027. On the facet-coating line we're planning roughly 10X capacity from base by early 2027, with a path to double again. The served-market framing of about $700 million by 2030 is management's forward view, not a booked result.
Give you some color on the concentration line. Our largest customer keeps growing as a share of revenue: under 10% two years ago, 11% in 2024, and 17% in 2025, and no other customer is above 10%. We don’t name the customer; the pattern – Taiwanese, semiconductor, advanced packaging – is consistent with a large foundry, but that’s inference, not disclosure.
Give you some color on the more than $250 million of orders that span all three lines, but the strength is uneven, and we say so. There are three pieces we serve in indium phosphide laser making: the epitaxy step that grows the active layer, the wet‑processing and etch steps, and the ion‑beam coating of the laser facets. On facet coating we’re a strong incumbent at a number of key customers; on epitaxy a competitor holds a good incumbent position and we are today the second source; on wet processing we hold a strong position with several of the leaders.
Give you some color on the more than $250 million of orders for indium phosphide laser tools — MOCVD epitaxy, wet processing, ion-beam facet coating — deliveries starting 2026, accelerating 2027. The driver: data centers shifting copper to optics, and a major AI-chip company investing in optical-networking suppliers. A large portion is facet-coating tools for 800G and 1.6T transceiver suppliers. We're planning roughly 10X capacity on that line by early 2027, with a path to double again. Management frames the served market at about $700 million by 2030 — a forward framing, not a booked result.
The more than $250 million of orders spans all three lines — epitaxy growing the active layer, wet-processing and etch, and ion-beam facet coating — but the strength is uneven, and we say so. Asked whether we're primary or second source, we split it by step: on facet coating we're a strong incumbent at a number of key customers; on epitaxy a competitor holds a good incumbent position and we're the second source, having built up products over recent years; on wet processing we hold a strong position with several leaders.
Give you some color on the quarter. Q1 came in at $158.3 million, down about 5% year over year at 36% non-GAAP gross margin — a touch below the midpoint of the guide and below the margin range. One laser-annealing system for a China customer needed an export license we were informed of by U.S. authorities, worth about $8 million. Q2 guided $170-190 million at 38-40% margin. Full year '26 still $740-800 million, growth weighted to the second half. Backlog ended '25 at $555 million, up about 35%, and management points to that as the visibility behind the second-half acceleration.
Give you some color on the concentration line. Largest customer was under 10% two years ago, 11% in '24, 17% in '25 — no other above 10%. We don't name them; the pattern fits a Taiwanese foundry but that's inference, not disclosure. Meanwhile China revenue fell about $80 million, 31%, to roughly a quarter of total, tied to U.S. export controls. With manufacturing about 99% U.S.-based, we're on the export-control side — restricted on what we can ship in. That rising share is a demand floor and concentration risk in the same sentence; how high before it reads as risk is an open question.
