diff --git a/.JuliaFormatter.toml b/.JuliaFormatter.toml
new file mode 100644
index 0000000000..c7439503e1
--- /dev/null
+++ b/.JuliaFormatter.toml
@@ -0,0 +1 @@
+style = "blue"
\ No newline at end of file
diff --git a/.github/workflows/Format.yml b/.github/workflows/Format.yml
new file mode 100644
index 0000000000..3f39a3fd32
--- /dev/null
+++ b/.github/workflows/Format.yml
@@ -0,0 +1,26 @@
+name: Format suggestions
+
+on:
+  pull_request:
+  
+concurrency:
+  # Skip intermediate builds: always.
+  # Cancel intermediate builds: only if it is a pull request build.
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
+
+jobs:
+  format:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: julia-actions/setup-julia@v1
+        with:
+          version: 1
+      - run: |
+          julia  -e 'using Pkg; Pkg.add("JuliaFormatter")'
+          julia  -e 'using JuliaFormatter; format("."; verbose=true)'
+      - uses: reviewdog/action-suggester@v1
+        with:
+          tool_name: JuliaFormatter
+          fail_on_error: true
diff --git a/lib/EnzymeCore/Project.toml b/lib/EnzymeCore/Project.toml
index ad6e42e371..f8f3721aad 100644
--- a/lib/EnzymeCore/Project.toml
+++ b/lib/EnzymeCore/Project.toml
@@ -1,7 +1,7 @@
 name = "EnzymeCore"
 uuid = "f151be2c-9106-41f4-ab19-57ee4f262869"
 authors = ["William Moses <wmoses@mit.edu>", "Valentin Churavy <vchuravy@mit.edu>"]
-version = "0.7.0"
+version = "0.7.1"
 
 [compat]
 Adapt = "3, 4"
diff --git a/lib/EnzymeCore/src/EnzymeCore.jl b/lib/EnzymeCore/src/EnzymeCore.jl
index 56c200cd61..680e9e42d3 100644
--- a/lib/EnzymeCore/src/EnzymeCore.jl
+++ b/lib/EnzymeCore/src/EnzymeCore.jl
@@ -239,6 +239,22 @@ end
 
 function tape_type end
 
+"""
+    compiler_job_from_backend(::KernelAbstractions.Backend, F::Type, TT:Type)::GPUCompiler.CompilerJob
+
+Returns a GPUCompiler CompilerJob from a backend as specified by the first argument to the function.
+
+For example, in CUDA one would do:
+
+```julia
+function EnzymeCore.compiler_job_from_backend(::CUDABackend, @nospecialize(F::Type), @nospecialize(TT::Type))
+    mi = GPUCompiler.methodinstance(F, TT)
+    return GPUCompiler.CompilerJob(mi, CUDA.compiler_config(CUDA.device()))
+end
+```
+"""
+function compiler_job_from_backend end
+
 include("rules.jl")
 
 end # module EnzymeCore
diff --git a/src/compiler.jl b/src/compiler.jl
index 676d1f302c..8405ccbdec 100644
--- a/src/compiler.jl
+++ b/src/compiler.jl
@@ -4758,7 +4758,7 @@ function GPUCompiler.codegen(output::Symbol, job::CompilerJob{<:EnzymeTarget};
         target_machine = GPUCompiler.llvm_machine(primal_job.config.target)
     end
 
-    parallel = Threads.nthreads() > 1
+    parallel = parent_job === nothing ? Threads.nthreads() > 1 : false
     process_module = false
     device_module = false
     if parent_job !== nothing