diff --git a/Makefile b/Makefile
index ad7f0ff3485f..1b4b21888a9c 100644
--- a/Makefile
+++ b/Makefile
@@ -99,7 +99,9 @@ endif
 
 ifeq ($(USE_TENSORRT), 1)
 	CFLAGS +=  -I$(ROOTDIR) -I$(TPARTYDIR) -DONNX_NAMESPACE=$(ONNX_NAMESPACE) -DMXNET_USE_TENSORRT=1
-	LDFLAGS += -lprotobuf -pthread -lonnx -lonnx_proto -lnvonnxparser -lnvonnxparser_runtime -lnvinfer -lnvinfer_plugin
+	LDFLAGS +=  -lprotobuf -pthread -lonnx -l:libonnx_proto.a -l:libnvonnxparser_static.a \
+	-l:libnvonnxparser_runtime_static.a -l:libnvonnxparser_plugin.a  -lnvinfer -lnvinfer_plugin \
+	-l:libonnx_proto.a -l:libonnxtrt_proto.a
 endif
 # -L/usr/local/lib
 
diff --git a/ci/docker/install/tensorrt.sh b/ci/docker/install/tensorrt.sh
index 641a88e5dc91..1950cad0b52f 100755
--- a/ci/docker/install/tensorrt.sh
+++ b/ci/docker/install/tensorrt.sh
@@ -26,7 +26,7 @@ pip3 install gluoncv==0.2.0
 pushd .
 cd ..
 apt-get update
-apt-get install -y automake libtool
+apt-get install -y automake libtool zip
 git clone --recursive -b 3.5.1.1 https://github.com/google/protobuf.git
 cd protobuf
 ./autogen.sh
diff --git a/ci/docker/runtime_functions.sh b/ci/docker/runtime_functions.sh
index 2ddbc84a8841..c564699c9716 100755
--- a/ci/docker/runtime_functions.sh
+++ b/ci/docker/runtime_functions.sh
@@ -79,6 +79,8 @@ build_wheel() {
     export MXNET_LIBRARY_PATH=${BUILD_DIR}/libmxnet.so
 
     cd ${PYTHON_DIR}
+
+    # If building for redistribution edit the name in this setup.py
     python setup.py bdist_wheel --universal
 
     # repackage
@@ -96,6 +98,8 @@ build_wheel() {
     rm -rf ${TMPDIR}
 
     popd
+
+    # If redistributing you may now run `twine upload -r pypi *.whl`
 }
 
 # Build commands: Every platform in docker/Dockerfile.build.<platform> should have a corresponding
@@ -498,7 +502,7 @@ build_ubuntu_gpu_tensorrt() {
     cd build
     cmake \
         -DCMAKE_CXX_FLAGS=-I/usr/include/python${PYVER}\
-        -DBUILD_SHARED_LIBS=ON ..\
+        -DBUILD_SHARED_LIBS=OFF ..\
         -G Ninja
     ninja -j 1 -v onnx/onnx.proto
     ninja -j 1 -v
@@ -514,12 +518,11 @@ build_ubuntu_gpu_tensorrt() {
     cmake ..
     make -j$(nproc)
     export LIBRARY_PATH=`pwd`:$LIBRARY_PATH
+    export LIBRARY_PATH=$LIBRARY_PATH:`pwd`/third_party/onnx/
+    mv third_party/onnx/libonnx_proto.a third_party/onnx/libonnxtrt_proto.a
     popd
 
     mkdir -p /work/mxnet/lib/
-    cp 3rdparty/onnx-tensorrt/third_party/onnx/build/*.so /work/mxnet/lib/
-    cp -L 3rdparty/onnx-tensorrt/build/libnvonnxparser_runtime.so.0 /work/mxnet/lib/
-    cp -L 3rdparty/onnx-tensorrt/build/libnvonnxparser.so.0 /work/mxnet/lib/
 
     rm -rf build
     make \
@@ -537,6 +540,8 @@ build_ubuntu_gpu_tensorrt() {
         ONNX_NAMESPACE=onnx                                  \
         CUDA_ARCH="-gencode arch=compute_70,code=compute_70" \
         -j$(nproc)
+
+    build_wheel /work/mxnet/python /work/mxnet/lib
 }
 
 build_ubuntu_gpu_mkldnn() {