diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 00000000..2d0c2c9e
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,55 @@
+FROM pytorch/pytorch:2.1.0-cuda12.1-cudnn8-runtime
+ENV DEBIAN_FRONTEND=noninteractive 
+
+ENV NVIDIA_VISIBLE_DEVICES=all NVIDIA_DRIVER_CAPABILITIES=all
+
+# Create a conda aloha env
+RUN conda create --name aloha python=3.8.10 && \
+    echo "source activate aloha" > ~/.bashrc
+
+# Set the PATH for the new environment
+ENV PATH /opt/conda/envs/aloha/bin:$PATH
+
+USER root
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    build-essential \
+    vim \ 
+    git
+
+RUN /opt/conda/envs/aloha/bin/python -m pip install \
+    torchvision \
+    torch \
+    pyquaternion \
+    pyyaml \
+    rospkg \
+    pexpect \
+    mujoco \
+    dm_control \
+    opencv-python \
+    matplotlib \
+    einops \
+    packaging \
+    h5py \
+    ipython
+
+# WORKDIR act-plus-plus/detr
+COPY detr /detr
+WORKDIR /detr
+RUN /opt/conda/envs/aloha/bin/python -m pip install -e .
+
+RUN apt-get update && apt-get install -y libglib2.0-0
+
+# Set the working directory
+WORKDIR /tempdir
+
+# Uninstall robomimic if it's installed
+RUN pip uninstall -y robomimic
+RUN git clone https://github.com/ARISE-Initiative/robomimic.git
+
+# Change into the robomimic directory and checkout the desired branch
+WORKDIR /tempdir/robomimic
+RUN git checkout diffusion-policy-mg
+
+# Install robomimic and diffusers
+RUN pip install -e . && pip install diffusers
+WORKDIR /act-plus-plus
\ No newline at end of file
diff --git a/README.md b/README.md
index 1e67d242..57f618b9 100644
--- a/README.md
+++ b/README.md
@@ -45,6 +45,14 @@ You can find all scripted/human demo for simulated environments [here](https://d
 
 - also need to install https://github.com/ARISE-Initiative/robomimic/tree/r2d2 (note the r2d2 branch) for Diffusion Policy by `pip install -e .`
 
+### Docker and Remote Development
+These instructions were written for both training and generating synthethic data on a remote box with a display head. In the event there is no head (ex: a GCP VM), follow the installation steps for setting up Chrome Remote Deskop. This will create a `display` which we can use below 
+1. In a terminal from a remote desktop screen, run `echo $DISPLAY` to get a display id (ex: `:20`)
+2. In a terminal started from your local machine, export the display id (ex: `export DISPLAY=:20`).
+3. In the same terminal, run `xhost+` to give containers access to the display for rendering.  
+4. `chmod +x run.sh && ./run.sh` to build a docker image and enter the container
+5. Additional volumes can be added for mapping in `run.sh`
+
 ### Example Usages
 
 To set up a new terminal, run:
diff --git a/run.sh b/run.sh
new file mode 100644
index 00000000..c8e3e211
--- /dev/null
+++ b/run.sh
@@ -0,0 +1,21 @@
+#!/bin/bash
+set -e
+set -u
+
+# Define the path to the volume you want to mount
+VOLUME_PATH="."
+
+# Build the Docker image. The -t option lets you tag your image so it's easier to find later.
+docker build -t aloha .
+
+# Check if the container exists
+if [ $(docker ps -a -f name=aloha | grep -w aloha | wc -l) -eq 0 ]; then
+    # Create the Docker container
+    docker run --name aloha -d -P -it -v /tmp/.X11-unix:/tmp/.X11-unix -v $VOLUME_PATH:/act-plus-plus --ipc=host --pid=host --network=host --gpus=all -w /act-plus-plus aloha:latest
+fi
+
+# Start the Docker container
+docker start aloha
+
+# Execute command in the Docker container
+docker exec -it aloha /bin/bash 
\ No newline at end of file