diff --git a/Include/Aurora/HWInfo/CpuLoadSampler.hpp b/Include/Aurora/HWInfo/CpuLoadSampler.hpp
index 46565567..4c198e4a 100644
--- a/Include/Aurora/HWInfo/CpuLoadSampler.hpp
+++ b/Include/Aurora/HWInfo/CpuLoadSampler.hpp
@@ -6,7 +6,8 @@
     Author: Reece
 
     Note: This API does not query system-wide CPU utilization
-          This belongs alongside AuProcess just about as much as our memory stat APIs do. I think it's fine to keep hardware perf query apis under AuHwInfo.
+          These APIs belong to the AuProcess subsystem just about as much as our memory-usage stat APIs do.
+           (I think it's fine to keep hardware perf query apis under AuHwInfo, even if they are process rel)
 ***/
 #pragma once
 
@@ -28,4 +29,5 @@ namespace Aurora::HWInfo
                          bool bCountKernelUsage = false);                    // False = Ignore kernel work under our callstack
 
     AUKN_SYM double GetProcessCPUUtilization();
+    AUKN_SYM double GetThreadCPUUtilization(AuOptional<bool> optIncludeKernel = { true });
 }
\ No newline at end of file
diff --git a/Source/HWInfo/AuCpuLoadSampler.cpp b/Source/HWInfo/AuCpuLoadSampler.cpp
index d8f4feff..403db5dc 100644
--- a/Source/HWInfo/AuCpuLoadSampler.cpp
+++ b/Source/HWInfo/AuCpuLoadSampler.cpp
@@ -10,6 +10,13 @@
 
 namespace Aurora::HWInfo
 {
+    static thread_local CpuLoadSamplerImpl tlsThreadLocalUsageSamplers[2] {
+        CpuLoadSamplerImpl(AuSToMS<AuUInt32>(1), true, false),
+        CpuLoadSamplerImpl(AuSToMS<AuUInt32>(1), true, true)
+    };
+
+    static CpuLoadSamplerImpl gProcessUsageSampler(AuSToMS<AuUInt32>(1), false, true);
+
     CpuLoadSamplerImpl::CpuLoadSamplerImpl(AuUInt32 uMinSamplePeriodMS,
                                            bool bThreadMode,
                                            bool bCountKernelUsage) :
@@ -112,8 +119,12 @@ namespace Aurora::HWInfo
 
     AUKN_SYM double GetProcessCPUUtilization()
     {
-        static CpuLoadSamplerImpl gSampler(AuSToMS<AuUInt32>(1), false, false);
-        return gSampler.GetLoad();
+        return gProcessUsageSampler.GetLoad();
+    }
+
+    AUKN_SYM double GetThreadCPUUtilization(AuOptional<bool> optIncludeKernel)
+    {
+        return tlsThreadLocalUsageSamplers[optIncludeKernel.value_or(true)].GetLoad();
     }
 
     AUKN_SYM ICpuLoadSampler *CpuLoadSamplerNew(AuUInt32 uMinSamplePeriodMS,