diff --git a/compile_paper.sh b/compile_paper.sh
new file mode 100755
index 00000000..81853164
--- /dev/null
+++ b/compile_paper.sh
@@ -0,0 +1,5 @@
+docker run --rm \
+    --volume $PWD:/data \
+    --user $(id -u):$(id -g) \
+    --env JOURNAL=joss \
+    openjournals/paperdraft
\ No newline at end of file
diff --git a/paper.md b/paper.md
new file mode 100644
index 00000000..7840f4b6
--- /dev/null
+++ b/paper.md
@@ -0,0 +1,68 @@
+---
+title: 'imodels: a python package for fitting interpretable models'
+tags:
+  - python
+  - machine learning
+  - interpretability
+  - explainability
+  - transparency
+  - decision rules
+authors:
+  - name: Chandan Singh^[Equal contribution]
+    orcid: 0000-0003-0318-2340
+    affiliation: 1 # (Multiple affiliations must be quoted)
+  - name: Keyan Nasseri^[Equal contribution]
+    affiliation: 1
+  - name: Bin Yu
+    affiliation: "1, 2"
+affiliations:
+ - name: EECS Department, University of California, Berkeley
+   index: 1
+ - name: Statistics Department, University of California, Berkeley
+   index: 2
+date: 27 January 2021
+bibliography: references.bib
+---
+
+# Summary
+
+`imodels` is a Python package for concise, transparent, and accurate predictive modeling.
+It provides users a simple interface for fitting and using state-of-the-art interpretable models, all compatible with scikit-learn [@pedregosa2011scikit].
+These models can often replace black-box models while improving interpretability and computational efficiency, all without sacrificing predictive accuracy.
+In addition, the package provides a framework for developing custom tools and rule-based models for interpretability.
+
+# Statement of need
+
+Recent advancements in machine learning have led to increasingly complex predictive models, often at the cost of interpretability.
+There is often a need for models which are inherently interpretable [@rudin2018please; @murdoch2019definitions], particularly in high-stakes applications such as medicine, biology, and political science.
+In these cases, interpretability can ensure that models behave reasonably, identify when models will make errors, and make the models more trusted by domain experts.
+Moreover, interpretable models tend to be much more computationally efficient then larger black-box models.
+
+Despite the development of many methods for fitting interpretable models [@molnar2020interpretable], implementations for such models are often difficult to find, use, and compare to one another.
+`imodels` aims to fill this gap by providing a simple unified interface and implementation for many state-of-the-art interpretable modeling techniques.
+
+# Features
+
+Interpretable models can take various forms.
+\autoref{fig:models} shows four possible forms a model in the `imodels` package can take.
+Each form constrains the final model in order to make it interpretable, but there are different methods for fitting the model which differ in their biases and computational costs.
+The `imodels` package contains implementations of various such methods and also useful functions for recombining and extending them.
+
+Rule sets consist of a set of rules which each act independently.
+    There are different strategies for deriving a rule set, such as Skope-rules [@skope] or Rulefit [@friedman2008predictive].
+Rule lists are composed of a set of rules which act in sequence, and include models such as Bayesian rule lists [@letham2015interpretable] or the oneR algorithm [@holte1993very].
+Rule trees are similar to rule lists, but allow branching after rules. This includes models such as CART decision trees [@breiman1984classification].
+Algebraic models take a final form of simple algebraic expressions, such as supersparse linear integer models [@ustun2016supersparse].
+
+![Examples of different supported model forms. The bottom of each box shows predictions of the corresponding model as a function of $X_1$ and $X_2$.\label{fig:models}](./docs/img/model_table.png){ width=100% }
+
+# Acknowledgements
+
+The code here heavily derives from the wonderful work of previous projects.
+In particular, we build upon the following repos and users: [sklearn-expertsys](https://github.com/tmadl/sklearn-expertsys) - by [Tamas Madl](https://github.com/tmadl) and [Benedict](https://github.com/kenben) based on original code by [Ben Letham](http://lethalletham.com/).
+We also based many rule-based models on [skope-rules](https://github.com/scikit-learn-contrib/skope-rules) by the [skope-rules team](https://github.com/scikit-learn-contrib/skope-rules/blob/master/AUTHORS.rst) (including [
+Nicolas Goix](https://github.com/ngoix), [Florian Gardin](https://github.com/floriangardin), [Jean-Matthieu Schertzer](https://github.com/datajms), Bibi Ndiaye, and Ronan Gautier). 
+We also build upon the [rulefit](https://github.com/christophM/rulefit) repository by [Christoph Molnar](https://github.com/christophM). 
+The authors would also like to acknowledge very useful feedback from Yan Shuo Tan during the early stages of this project.
+
+# References
\ No newline at end of file
diff --git a/paper.pdf b/paper.pdf
new file mode 100644
index 00000000..55818e20
Binary files /dev/null and b/paper.pdf differ
diff --git a/references.bib b/references.bib
new file mode 100644
index 00000000..282de3fa
--- /dev/null
+++ b/references.bib
@@ -0,0 +1,96 @@
+@article{ustun2016supersparse,
+  title={Supersparse linear integer models for optimized medical scoring systems},
+  author={Ustun, Berk and Rudin, Cynthia},
+  journal={Machine Learning},
+  volume={102},
+  number={3},
+  pages={349--391},
+  year={2016},
+  publisher={Springer}
+}
+
+@article{pedregosa2011scikit,
+  title={Scikit-learn: Machine learning in Python},
+  author={Pedregosa, Fabian and Varoquaux, Ga{\"e}l and Gramfort, Alexandre and Michel, Vincent and Thirion, Bertrand and Grisel, Olivier and Blondel, Mathieu and Prettenhofer, Peter and Weiss, Ron and Dubourg, Vincent and others},
+  journal={the Journal of machine Learning research},
+  volume={12},
+  pages={2825--2830},
+  year={2011},
+  publisher={JMLR. org}
+}
+
+@article{letham2015interpretable,
+  title={Interpretable classifiers using rules and bayesian analysis: Building a better stroke prediction model},
+  author={Letham, Benjamin and Rudin, Cynthia and McCormick, Tyler H and Madigan, David and others},
+  journal={Annals of Applied Statistics},
+  volume={9},
+  number={3},
+  pages={1350--1371},
+  year={2015},
+  publisher={Institute of Mathematical Statistics}
+}
+
+@article{holte1993very,
+  title={Very simple classification rules perform well on most commonly used datasets},
+  author={Holte, Robert C},
+  journal={Machine learning},
+  volume={11},
+  number={1},
+  pages={63--90},
+  year={1993},
+  publisher={Springer}
+}
+
+@book{breiman1984classification,
+  title={Classification and regression trees},
+  author={Breiman, Leo and Friedman, Jerome and Stone, Charles J and Olshen, Richard A},
+  year={1984},
+  publisher={CRC press}
+}
+
+@book{molnar2020interpretable,
+  title={Interpretable machine learning},
+  author={Molnar, Christoph},
+  year={2020},
+  publisher={Lulu. com}
+}
+
+@article{rudin2018please,
+  title={Please stop explaining black box models for high stakes decisions},
+  author={Rudin, Cynthia},
+  journal={arXiv preprint arXiv:1811.10154},
+  volume={1},
+  year={2018},
+  publisher={Nov}
+}
+
+@article{murdoch2019definitions,
+  title={Definitions, methods, and applications in interpretable machine learning},
+  author={Murdoch, W James and Singh, Chandan and Kumbier, Karl and Abbasi-Asl, Reza and Yu, Bin},
+  journal={Proceedings of the National Academy of Sciences},
+  volume={116},
+  number={44},
+  pages={22071--22080},
+  year={2019},
+  publisher={National Acad Sciences}
+}
+
+@article{friedman2008predictive,
+  title={Predictive learning via rule ensembles},
+  author={Friedman, Jerome H and Popescu, Bogdan E and others},
+  journal={Annals of Applied Statistics},
+  volume={2},
+  number={3},
+  pages={916--954},
+  year={2008},
+  publisher={Institute of Mathematical Statistics}
+}
+
+@misc{skope,
+  author = {{Skope Collaboration}},
+  title = {Skope-rules},
+  year = {2021},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  url = {https://github.com/scikit-learn-contrib/skope-rules}
+}
\ No newline at end of file