WorldIndustrialWeekly General: Test Limits Behind Kaynes Semicon, NI, and India's OSAT Scaling Outlook

For production, procurement, and engineering leaders using WorldIndustrialWeekly's cross-sector industrial baseline, the issue now is not whether assembly and test capacity appears on a map. It is whether test coverage, qualification timelines, and device-migration architecture can support output when five semiconductor plants in India move into commercial production. That is where NI and Kaynes Semicon intersect with the July 2026 SEMI forecast: test equipment sales are expected to rise 31% to $15.3 billion in 2026 after 55.3% growth in 2025, while $20.8 billion is projected for 2028.

The capacity announcement is not the test plan

A headline such as 15 million chips a year does not answer how many testers are required. Two products with identical unit volume can imply very different equipment counts if one is tested under a minute and another requires five to 10 minutes. The planning variable, therefore, is test cycle time at the device level, not just factory capacity.

Case record: Kaynes Semicon and the STS deployment

• In September 2025, Emerson announced a partnership with Kaynes Semicon to deploy the NI Semiconductor Test System (STS) as the primary test platform at Kaynes’ OSAT facility in Sanand, Gujarat.
• EE Times spoke with Karthik Balasekar, chief business manager for semiconductor and electronics, and Ved Mahanta, principal account manager, both at NI India, part of Emerson.

Qualification windows shape purchasing decisions

Mahanta said, “Customers typically conduct tester qualification before making a final purchasing decision,” and placed the evaluation window at six months to one year. He gave the public Kaynes example as follows: “The process generally takes six months to one year. In the case of Kaynes, we can publicly state that the company has decided to proceed with NI testers. We are in the process of delivering the testers in the coming months.”

Mahanta added that negotiations with other OSAT sites remained active, while NI could not identify additional companies because some engagements were not disclosed publicly.

Line-level output does not equal tester demand

Balasekar linked tester count to the product mix running through the line. “Facilities handling millions of devices may need fewer testers if each test takes less than a minute,” he said. “Testing complex chips that require five or 10 minutes per device could demand more equipment to meet the same production target.”

For less complex chips, pin count and required tests still shape throughput. A device with fewer pins and lighter test requirements can consume less tester time, letting the same equipment process more units.

On a production line, the goal is to test as many chips as possible in the shortest time while confirming specification compliance. TestStand, NI’s test automation platform, automates sequences so batch disposition decisions arrive faster, Mahanta said.

Mahanta also cautioned against reading capacity statements as equipment counts: “An OSAT might announce an annual capacity of 14 million or 15 million chips, for example. Such figures generally represent the combined output capacity of its production lines, whether the facility operates 10 lines, 50 lines, or another number.” He said NI Emerson had not disclosed a Kaynes test-system quantity and compared OSAT announcements to automotive output claims. “Actual production depends on customer demand,” he noted, with capacity alone not guaranteeing volume.

Device mix determines platform usefulness

For Kaynes, the choice was described by Balasekar as a platform decision aimed at handling different semiconductor device types. “Our STS is a platform that supports both mixed-signal and RF applications,” he said. “For example, if Kaynes tests mixed-signal products today and moves to RF products later, it can reuse the same STS tester for different devices under test.”

Legacy migration and new-product development require different preparation

Mahanta separated customer needs into two classes: “Typically, customers want to test two types of devices: legacy devices and new products, commonly referred to as new product development,” he said. Legacy flows may already run on platforms whose throughput no longer meets requirements. “OSATs need to move these products to newer systems without compromising test coverage,” he added, while customers often complete much of the preparation and specify what the OSAT must test.

Startups add a further gap. “Some Indian startups developing their own chips have never performed final testing and depend on the OSAT to guide them,” Mahanta said. “The OSAT must help determine how to test each device reliably and quickly enough to meet production and time-to-market requirements.”

Even without legacy technical debt, a new device can force architecture work from the ground up. “This includes determining the test requirements, selecting an appropriate system, and ensuring that it meets test coverage and throughput targets,” Mahanta said. He noted that the chosen supplier may not be predetermined: “Such a system could come from NI Emerson or another test equipment supplier.”

Balasekar said both workloads become harder as design complexity rises.

PXI modularity as a hedge against changing test content

When a device adds features, the OSAT needs systems that can absorb new requirements while preserving coverage. Mahanta pointed to modular PXI (PCI eXtensions for Instrumentation) approaches, where additional modules can be introduced without altering existing test parameters. “Customers can expand their testing capabilities as their requirements evolve,” he said.

AI-assisted architecture: a starting point, not an engineer replacement

For new devices, NI uses AI in portions of test development, including early architecture design. Mahanta described an engineer uploading a chip datasheet to NI’s Nigel AI test development tool; the system reviews selected parameters and suggests a possible architecture. “The engineer could then use the suggested framework as a starting point rather than developing the entire architecture from scratch,” he said.

The limitation is explicit. “This approach targets part of the work involved in test development, rather than replacing the validation engineer,” Mahanta said. “AI could take on some repetitive tasks, leaving engineers to focus on more complex problem-solving.”

ISM 2.0, engineering depth, and workforce pressure

Under India Semiconductor Mission 2.0 (ISM 2.0), Ashwini Vaishnaw said at SEMICON India 2026 that government attention will cover six areas, including advanced packaging, testing units, and talent. For OSAT operators, that means test system flexibility, device qualification, and accommodation of different device requirements remain central to the next expansion phase.

Mahanta said NI has built local engineering support so OSAT customers can be served on site, lowering the need to fly specialists from Taiwan or Singapore. He also mentioned Tessolve, an NI partner serving global customers, as an Indian firm in the testing ecosystem.

“There are integrated device manufacturers that base 40% to 50% of their product lines in India,” Mahanta said. “Building engineering expertise will be essential for Indian companies seeking to serve global semiconductor manufacturers.”

Balasekar said growth in OSATs and electronics manufacturing may outrun skilled-worker supply, given India’s electronic device exports and expanding mobile phone manufacturing base. He pointed to ISM 2.0 workforce development as a way to narrow the gap: “I think that will help India grow faster and keep pace with the market’s expansion,” he said.

Common misconceptions in OSAT test planning

Misconception	Practical implication drawn from the source discussion
Annual chip volume alone determines tester count.	Cycle time per device is decisive; simpler and complex products can need different tester counts at the same target output.
Test-technology selection can happen immediately before shipment.	Qualification is a procurement gate and can take a planning window of half a year to a year.
Flexible hardware makes all architecture work unnecessary.	Legacy migration and startup products still require coverage-preserving migration, test-requirement definition, and throughput validation.
AI can independently build production test programs.	AI is described as assisting early architecture work while validation engineering remains necessary.

Action recommendations

1. Model tester demand from cycle time. Separate device families by short-run and long-run test profiles before converting capacity targets into tester purchases.
2. Reserve time for tester qualification. Procurement schedules should acknowledge that qualification can be a multi-month decision gate, not a final paperwork step.
3. Match platform choice to future device mix. Evaluate whether a test system can support different device classes, such as mixed-signal and RF, without rebuilding the program set.
4. Plan legacy migration around coverage. When moving established products to newer platforms, preserve coverage while closing throughput gaps.
5. Use modular architecture where requirements are uncertain. Systems with modular expansion can reduce rework when new tests appear.
6. Apply AI at the architecture-start stage. Use AI-assisted tools to create candidate starting points, while keeping validation engineers responsible for production readiness.
7. Track policy programs for skills. National semiconductor programs include workforce development, which is relevant to test engineering staffing.

FAQ

Q: Why does the story concern production managers, not only equipment buyers?

Because assembly and test sites can reach commercial output only if test coverage, cycle time, qualification, and workforce capacity are solved together. The case of Kaynes and NI shows that the tester decision is embedded in production targets.

Q: What is the most misunderstood capacity number?

An annual chip figure can represent combined line capacity and may not disclose tester needs. Actual output also depends on customer demand.

Q: What role does PXI play?

PXI is described as modular, allowing test capabilities to expand with evolving requirements without changing existing parameters.

Q: What does ISM 2.0 signal for testing?

Government focus spans six areas, including packaging, test units, and talent, making test architecture flexibility and engineering supply relevant to growth.

The operating lesson is that test capacity is a system problem: tester count, qualification time, architecture migration, modular expansion, AI-assisted design, and talent planning all interact with production targets.