2026-06-24XQuote

Source item

OpenLight (private) seems to get bigger and bigger every time I look at it. If you’re curious about their public ecosystem outside of Advantest: $JBL ( $SIVE partner ) for scaling PICs $MRVL and $MXL for DSPs $TSEM for the foundry. $300394.SZ / TFC Optical (OSAT/subassembly) Can often get exposure into private growth through public equities if you’re not an accredited investor. Since optical players look very interconnected.

Quoted context

These are the quoted author’s words.

Advantest (TSE: 6857 / $ATEYY) × OpenLight: Silicon Photonics HVM Test Partnership Advantest and OpenLight have announced a partnership to co-develop scalable, high-volume manufacturing (HVM) test solutions for silicon photonics (SiPh) devices. The demand logic runs cleanly through the release: expanding AI/HPC workloads drive data-center adoption of SiPh and co-packaged optics (CPO), which in turn surfaces a production-stage test bottleneck. That bottleneck is the high-volume testing of electro-optical (EO) devices, specifically the Optical Engine. SiPh devices demand simultaneous electrical measurement and optical coupling/alignment, so conventional electrical ATE infrastructure alone cannot deliver the throughput needed to scale. The two companies target an end-to-end test framework spanning wafer sort through final test, with OpenLight contributing its III-V heterogeneous integration PASIC (Photonic ASIC) and PDK, and the release citing a recent $50M Series A-1 round (total funding $84M) as evidence of commercial momentum. Advantest enters SiPh test directly, sharpening the ATE duopoly contest. This move puts Advantest in head-to-head positioning against Teradyne's Photon 100 in the EO/tester layer, with both incumbents now moving to absorb production-test demand from fabless PIC design houses like OpenLight. The constraint is structural, not incremental. Combining electrical test with optical alignment at volume is the hard part; framing it as "end-to-end, wafer sort to final test" signals the partners are targeting the full insertion chain rather than a single station, which is where throughput economics actually break. Read against the "Three Pillars" frame, this is an ATE-side consolidation signal. It complements rather than displaces the AEHR/FORM/KEYS wafer-test framework (https://t.co/XvOhml7Zok), but the specific division of labor (optical alignment method, WLBI applicability, insertion architecture) is unstated and rests on a promotional first-party release, so it warrants cross-checking against later conference disclosures.
PhotonCap ↗