Sixsense

SixSense (legal name SixSense Pte. Ltd) is a Singapore-headquartered deep tech company that develops an AI platform for semiconductor manufacturing, covering automated defect inspection, lot disposition, root-cause analysis and defect prediction. Its stated purpose is to convert raw production data — defect images and equipment signals — into real-time insight so that fabs can prevent quality escapes and improve yield rather than relying on engineers to manually interpret dashboards, SPC charts and inline inspection output.

The product line comprises three flagship components. AI-ADC (also referred to as classifAI) is a real-time vision engine that analyses captured inspection images and classifies suspected defects into hundreds of categories, with company-stated accuracy above 99% and no escapees; it includes tooling for labelling, model training, testing, deployment and maintenance, and runs 24/7 inferencing at line speed. Lot Disposition and Root-Cause Analysis builds on those labels to decide pass, hold, rework or scrap for each lot based on defect severity, recurrence patterns and yield impact, and correlates classified defects against tool, recipe and chamber data across hundreds of process steps to flag parameters drifting out of spec. Predict AI forecasts when, where and what to inspect by learning defect likelihood across time, layers and lots, triggering proactive changes to inspection sampling and tools.

Earlier public material describes classifAI as a GUI-based platform aimed at defect and yield engineers, with an intelligent annotation module that uses AI to accelerate labelling and an analytics module that aggregates defect locations into heatmaps to expose systematic failure patterns. The platform has been applied across 100mm–300mm wafer standards, on both pure silicon and compound wafers, and for end markets including RF and automotive. It is offered cloud-first with on-premise deployment available.

Founding story

SixSense was founded in 2018 by Akanksha Jagwani (CEO) and Avni Agrawal (CTO), both engineers. Jagwani studied mechanical engineering at IIT Gandhinagar and previously held engineering roles at Altair and built automation solutions and led product work at companies including Hyundai Motors, GE, Embibe and Headout. Agrawal, a computer science graduate of Motilal Nehru National Institute of Technology, previously worked at The D. E. Shaw Group and Visa, where she built large-scale data analytics systems, some of which were later protected as trade secrets. The pair evaluated sectors from aviation to automotive before choosing semiconductors, and after speaking with more than 50 engineers concluded that inspection processes in chip manufacturing remained largely manual and fragmented despite the industry's precision requirements.

Business model

Software platform sold to semiconductor manufacturers (foundries, OSATs and IDMs) and deployed either in the cloud or on-premise, integrating with existing AOI and inspection equipment rather than requiring new capital equipment. Sources do not disclose pricing or contract structure.

Not disclosed in the available sources; the company sells an enterprise AI software platform to chip manufacturers.

Traction

Named customers include GlobalFoundries, JCET and Infineon Technologies, with DenseLight Semiconductor's CEO also quoted. The company reports more than 100 million chips processed to date, hundreds of millions of images classified per year, and customer-reported outcomes of up to 30% faster production cycles, a 1–2% yield improvement and a 90% reduction in manual inspection work. A startup directory lists headcount at 41–60.

Latest developments

In July–August 2025 SixSense announced an $8.5 million Series A led by Peak XV's Surge with participation from Alpha Intelligence Capital and FEBE, taking total funding to about $12 million, and stated it is expanding from Asian fabs into the U.S. market. Its website markets an 'autonomous fab' positioning built around AI-ADC, Lot Disposition and Root-Cause Analysis, and Predict AI.