Designing custom mobile processors is a notoriously slow, painstaking process. Yet, Samsung’s System LSI division—the business unit responsible for Exynos processors—has managed to drastically shorten its development cycles by deploying Anthropic’s AI coding tool, Claude Code, directly into its semiconductor workflows.
According to a report from South Korean outlet Chosun Biz, Samsung first opened Claude Code to software developers in May before expanding its use to specialized hardware verification and design. The results have been dramatic: in one custom system-on-chip (SoC) project featuring 64 complex intertwined data paths, Claude set up a virtual test environment and ran verification scenarios in just two days. This process was internally assessed as 15 times faster than traditional methods that normally take over a month.
Even more impressively, the AI handled this task despite missing design files. When standard documentation and the Register Transfer Level (RTL) code for a DRAM controller were delayed, engineers fed Claude available SoC specs and electronic design automation (EDA) vendor data. The AI inserted virtual placeholder blocks to inspect core data paths, catching errors before even finishing the real circuit design.
Flattening learning curves and narrowing the Qualcomm gap
In another case, a second-year engineer with no prior experience in vibe coding or Claude Code used the AI tool to construct virtual USB keyboard and mouse models for an emulator. The assignment typically requires weeks of studying communication standards and adapting reference code. But thanks to Claude, Samsung finished it in one day. The tool allowed the engineer to develop an Android OS USB device driver.
Narrowing the workforce gap with Qualcomm
These leaps in efficiency are crucial for Samsung right now. LSI’s System division has around 6,000 employes, compared to the roughly 52,000 employes—almost nine times as many—at its main rival in the mobile application processor market, Qualcomm.
After reporting losses in its SoC business and seeing flagships like the Galaxy Z Fold 8 rely entirely on Qualcomm Snapdragon chips, Samsung is using AI to automate repetitive tasks and help junior staff develop their skills faster.
This shift is part of Samsung’s group-wide “Great AI Transformation” push. The strategy integrates external generative tools—including Google Gemini, OpenAI’s ChatGPT, and Claude—across research, manufacturing, marketing, and support. Similar AI applications in Samsung’s Memory division have already cut process design kit recalibrations by over 95%.
Why human engineers are still in charge
Despite these productivity gains, Samsung is far from putting its chip design on autopilot. Claude Code has made several notable blunders that highlight its lack of deep hardware context and understanding of hardware description languages.
In one example, the AI needed to fix an error. Instead of fixing the root cause, the AI just swapped the system alert from a “error” message to a casual “information” tag. In other cases, Claude accidentally restored unrelated, finished code while attempting to undo a single feature. The AI even tried to modify core RTL circuit designs it wasn’t authorized to touch.
Physical semiconductor flaws cannot be patched with a quick software update once mass production starts. So, Samsung treats Claude strictly as an assistant. Human engineers define the project boundaries, set the goals, and thoroughly re-verify every line of code to prevent costly hardware disasters.
The post Samsung Uses Claude AI to Slash Month-Long Chip Design Tasks to Just Days, But Humans Remain Key appeared first on Android Headlines.
