diff --git a/examples/hello_world/Makefile b/examples/hello_world/Makefile
new file mode 100644
index 0000000..fc0ffdf
--- /dev/null
+++ b/examples/hello_world/Makefile
@@ -0,0 +1,19 @@
+F90=nvfortran
+F90FLAGS=-cuda
+
+.PHONY: clean distclean
+
+all:  hello_world_cpu.x hello_world_cuda.x mem_alloc.x
+
+hello_world_cuda.x: hello_world_cuda.f90
+hello_world_cpu.x: hello_world_cpu.f90
+mem_alloc.x: mem_alloc.f90
+
+%.x: %.f90
+	$(F90) $(F90FLAGS) $< -o $@
+
+distclean: clean
+	rm *.x
+
+clean:
+	rm -f *.o *.mod
diff --git a/examples/hello_world/hello_world_cpu.f90 b/examples/hello_world/hello_world_cpu.f90
new file mode 100644
index 0000000..27e3550
--- /dev/null
+++ b/examples/hello_world/hello_world_cpu.f90
@@ -0,0 +1,17 @@
+module helloWorld
+  implicit none
+
+contains
+
+  subroutine hello_world_cpu
+    write(*,*) 'hello world'
+  end subroutine hello_world_cpu
+end module helloWorld
+
+program testHelloWorld
+  use helloWorld
+
+  implicit none
+
+  call hello_world_cpu
+end program testHelloWorld
diff --git a/examples/hello_world/hello_world_cuda.f90 b/examples/hello_world/hello_world_cuda.f90
new file mode 100644
index 0000000..c77a1c2
--- /dev/null
+++ b/examples/hello_world/hello_world_cuda.f90
@@ -0,0 +1,32 @@
+module helloWorld
+  implicit none
+
+contains
+
+  subroutine hello_world_cpu
+    write(*,*) 'hello world from CPU code'
+  end subroutine hello_world_cpu
+
+  attributes(global) subroutine hello_world_cuda
+    write(*,*) 'hello world from CUDA code'
+  end subroutine hello_world_cuda
+
+  attributes(global) subroutine hello_world_cuda_threads
+    write(*,*) 'hello world from thread', threadIdx%x
+  end subroutine hello_world_cuda_threads
+end module helloWorld
+
+program testHelloWorld
+  use cudafor
+  use helloWorld
+
+  implicit none
+
+  integer :: istat
+
+  call hello_world_cpu
+  call hello_world_cuda<<<1, 1>>>
+  istat = cudaDeviceSynchronize() 
+  call hello_world_cuda_threads<<<1, 8>>>
+  istat = cudaDeviceSynchronize() 
+end program testHelloWorld
diff --git a/examples/hello_world/mem_alloc.f90 b/examples/hello_world/mem_alloc.f90
new file mode 100644
index 0000000..2d9c3c9
--- /dev/null
+++ b/examples/hello_world/mem_alloc.f90
@@ -0,0 +1,41 @@
+module memoryTest
+  implicit none
+
+contains
+
+  subroutine print_cpu(a)
+    real, intent(in) :: a(:)
+    write(*,*) a
+  end subroutine print_cpu
+
+
+  attributes(global) subroutine modify_from_gpu(a)
+    real, intent(out) :: a(:)
+    a = a + 10.0
+  end subroutine modify_from_gpu
+end module memoryTest
+
+program testMemory
+  use cudafor
+  use memoryTest
+
+  implicit none
+
+  integer :: istat, i, j, N
+  real, managed, allocatable :: a(:)
+
+  N = 8
+  allocate(a(N))
+
+  do i = 1, N
+     a(i) = i * 0.124
+  end do
+
+  write(*,*) 'Before GPU modification'
+  call print_cpu(a)
+  call modify_from_gpu<<<1, 1>>>(a)
+  istat = cudaDeviceSynchronize() 
+  write(*,*) 'After GPU modification'
+  call print_cpu(a)
+  deallocate(A)
+end program testMemory
diff --git a/examples/hello_world/script.sh b/examples/hello_world/script.sh
new file mode 100644
index 0000000..0188c13
--- /dev/null
+++ b/examples/hello_world/script.sh
@@ -0,0 +1,16 @@
+#!/bin/bash -l
+#SBATCH --nodes=1
+#SBATCH --ntasks-per-node=1
+#SBATCH --ntasks-per-core=1
+#SBATCH --cpus-per-task=1
+#SBATCH --gres=gpu:1
+#SBATCH --reservation=spc-cuda-training-12.04
+#SBATCH --account=spc-cuda-training
+#SBATCH --time=0:05:00
+
+module load nvhpc
+
+srun  -n 1 ./hello_world_cuda.x > output_hello_world_cpu
+srun  -n 1 ./hello_world_cuda.x > output_hello_world_cuda
+srun  -n 1 ./mem_alloc.x > output_mem_alloc
+