---
title: "Wooptix Targets AI Packaging Bottleneck with Astronomy Tech"
type: "News"
locale: "en"
url: "https://longbridge.com/en/news/285914026.md"
description: "Wooptix, a Spain-based startup, aims to address manufacturing bottlenecks in AI processor packaging using wavefront-sensing technology originally developed for astronomy. Their method allows for rapid and detailed wafer inspections, crucial as chipmakers shift towards advanced packaging techniques. CEO José Ramos emphasizes the company's focus on semiconductors, citing the industry's potential. While currently suited for research and development environments, the challenge remains to penetrate full-scale production fabs, where consistent monitoring and yield improvement are essential."
datetime: "2026-05-11T09:25:47.000Z"
locales:
  - [zh-CN](https://longbridge.com/zh-CN/news/285914026.md)
  - [en](https://longbridge.com/en/news/285914026.md)
  - [zh-HK](https://longbridge.com/zh-HK/news/285914026.md)
---

# Wooptix Targets AI Packaging Bottleneck with Astronomy Tech

//php echo do\_shortcode('\[responsivevoice\_button voice="US English Male" buttontext="Listen to Post"\]') ?\>

As chipmakers race to build faster AI processors, attention has shifted beyond transistor scaling and toward advanced packaging, hybrid bonding, and 3D integration. That shift is creating new manufacturing bottlenecks—among them, how to precisely measure wafer shape, topography, and warpage as devices become thinner, denser, and increasingly stacked.

One company hoping to capitalize on that transition is Wooptix, a Spain-based startup whose core technology originated not in semiconductor fabs, but in astronomy labs in the Canary Islands.

The company says wavefront-sensing techniques originally developed to correct atmospheric distortion in telescopes can now help semiconductor manufacturers inspect wafers faster and with greater detail than some conventional approaches.

The bigger question is whether that scientific promise can translate into something far harder: becoming a trusted supplier inside one of the world’s most conservative manufacturing industries.

### **From adaptive optics to wafer metrology**

CEO José Ramos traces the company’s roots to adaptive optics, where astronomers compensate for atmospheric turbulence that distorts incoming light from stars.

“The atmosphere changes every 10 milliseconds,” Ramos told EE Times. “If you can measure those changes and correct them in real time, you see the star properly again.”

Traditional astronomy systems often relied on Shack-Hartmann sensors, but Ramos said Wooptix developed a higher-resolution way to reconstruct wavefront phase information using simpler optical hardware and proprietary algorithms.

That expertise eventually found a new target: silicon wafers.

“At the moment we understood wafer warpage could be obtained in just one shot with our technology, we knew this was a major opportunity,” Ramos said. Wooptix ultimately chose semiconductors over other markets, including medical imaging and automotive.

“Semiconductor is the best business in the world,” Ramos said. “And the company needed focus.”

Moreover, AI workloads are changing. Rather than relying solely on transistor scaling, chipmakers increasingly use chiplets, stacked memory, advanced interposers, and heterogeneous integration to improve performance and energy efficiency.

According to Carlos Beitia, a metrology expert at CEA-Leti who previously worked in semiconductor metrology at KLA, that makes package-level mechanics strategically important. “We are seeing a clear trend today in the industry,” he told EE Times. “Everybody is looking to the package.”

As multiple materials are bonded together, each with different coefficients of thermal expansion, controlling deformation becomes more difficult. That creates demand for tools capable of measuring full-wafer topography quickly and at increasingly fine resolution.

Wooptix argues its approach can gather large amounts of surface information in a single capture rather than through slower scanning methods. Instead of scanning the wafer point by point, the system captures how light waves are altered as they reflect from the surface, using that phase information to reconstruct the wafer’s shape in a single measurement. This allows full-field topography to be obtained in milliseconds rather than through sequential sampling.

Ramos said existing interferometry-based approaches can be effective, but often involve tradeoffs in throughput or resolution. He claimed Wooptix can generate millions of measurement points in milliseconds while preserving high spatial detail.

Beitia said the company’s potential advantages extend beyond speed. “Their technology can gather a huge amount of data in seconds,” he said.

He also highlighted the company’s ability to work with both intensity and phase information, which could help in difficult multilayer structures where signal interpretation becomes more complex. Perhaps most interestingly, Beitia said Wooptix’s astronomy heritage may offer benefits in extracting weak signals from noisy environments.

“One of the major issues astronomy has is how do you make measurements with just a few photons,” he said.

### **Where Wooptix fits today—and where it needs to go**

For now, Wooptix appears best positioned in research environments, pilot lines, and process-development settings rather than full-scale production fabs. Beitia said one immediate use case is helping research institutes and R&D organizations measure wafer flatness or topography before and after process steps to understand how materials and manufacturing operations change the substrate.

“You measure once before, you do your experiments, then you measure how flat it is again,” he said, describing a typical development workflow.

That kind of metrology can be especially valuable in advanced packaging research, where bonding, deposition, CMP, thermal cycling, and heterogeneous materials can introduce subtle deformation that later affects yield or alignment.

In such environments, the commercial equation is different from high-volume manufacturing. A small number of tools may be enough if they provide insight that speeds development or helps engineers debug difficult processes. But Beitia drew a clear distinction between R&D adoption and strategic scale.

“If you’re talking about research, or even fabs, you’re talking about two tools max per site,” he said. “If you’re talking about process control, then you’re talking about large-scale business.”

That’s why the real opportunity—and the real challenge—lies inside production fabs, where tools are deployed repeatedly across lines to monitor wafers, catch outliers, and support yield improvement.

But reaching that stage is far harder. Equipment suppliers must prove uptime, repeatability, automation compatibility, service capability, and clear return on investment—often through lengthy customer evaluations.

Ramos acknowledged that industrialization is his company’s next hurdle. “The large incumbents are huge,” he said. “They have a reputation. They have customers. They have broad teams for 24-7 maintenance. We just have technology.”

Beitia noted that the industry has, at times, helped smaller companies industrialize valuable technologies when they solve urgent problems. He pointed to examples such as Nearfield Instruments, which gained backing from major chipmakers including TSMC as fabs sought new metrology capabilities not available from established suppliers.

That history suggests startups with differentiated tools can still find pathways into fabs—if they solve a painful enough problem.

### **Europe’s missing layer and Spain’s deep-tech bet**

The Wooptix story also highlights a broader European challenge. Europe has world-class semiconductor assets in areas such as lithography, automotive chips, power electronics, and R&D institutes. But it has produced relatively few scaled metrology and process-control champions.

Beitia believes more is needed and pointed out that global competitors increasingly consolidate technologies while Europe often fragments effort across too many small initiatives. “In Europe, we are not consolidating; we are spreading,” he said.

That fragmentation can make it harder for promising sensor or algorithm companies to become durable equipment businesses.

Wooptix is also part of a growing Spanish semiconductor ecosystem that includes fabless startups, photonics initiatives, public R&D centers, and increasing government support.

Emilio Garcia, a semiconductor policy expert and former Spanish government official, described Wooptix as one of Spain’s standout semiconductor bets. “Wooptix is certainly a company to watch,” he told EE Times.

Spain has recently sought to strengthen domestic capabilities through funding programs and strategic investments tied to broader digital sovereignty goals, creating a more supportive environment for deep-tech firms than existed a decade ago.

For Wooptix, the next phase is less about theory than execution. Ramos said success would mean emerging from evaluation cycles with customers ready to order systems in volume.

“At that moment, we will be a new eye in the field,” he said, suggesting Wooptix could offer chipmakers a new way to observe and control wafer behavior in advanced manufacturing.

That ambition is bold. But the market timing may be better than it first appears. As AI pushes semiconductor manufacturing into a new era of 3D integration and packaging complexity, the industry may be more open than before to specialized new entrants that solve specific pain points.

Whether Wooptix becomes one of them will depend not on where the technology came from, but whether it can survive where it is trying to go.

* * *

##### Read also:

* * *

### Related Stocks

- [588170.CN](https://longbridge.com/en/quote/588170.CN.md)
- [PSI.US](https://longbridge.com/en/quote/PSI.US.md)
- [159325.CN](https://longbridge.com/en/quote/159325.CN.md)
- [XSD.US](https://longbridge.com/en/quote/XSD.US.md)
- [SMH.US](https://longbridge.com/en/quote/SMH.US.md)
- [512480.CN](https://longbridge.com/en/quote/512480.CN.md)
- [512760.CN](https://longbridge.com/en/quote/512760.CN.md)
- [512720.CN](https://longbridge.com/en/quote/512720.CN.md)
- [SOXL.US](https://longbridge.com/en/quote/SOXL.US.md)
- [588780.CN](https://longbridge.com/en/quote/588780.CN.md)
- [159995.CN](https://longbridge.com/en/quote/159995.CN.md)
- [SOXX.US](https://longbridge.com/en/quote/SOXX.US.md)
- [KLAC.US](https://longbridge.com/en/quote/KLAC.US.md)

## Related News & Research

- [China's CXMT expects revenue to surge as memory chip demand soars](https://longbridge.com/en/news/286706506.md)
- [Where Cerebras' monster debut puts in among tech's biggest IPOs](https://longbridge.com/en/news/286465752.md)
- [Tower signs customer contracts for $1.3bn silicon photonics revenue for 2027](https://longbridge.com/en/news/286379624.md)
- [KLA approves 10-for-1 stock split and dividend hike](https://longbridge.com/en/news/286340025.md)
- [Tower Semiconductor Analysts Boost Their Forecasts Following Better-Than-Expected Q1 Earnings](https://longbridge.com/en/news/286435168.md)