Atanas Pavlov, Developer in Gabrovo, Bulgaria
Atanas is available for hire
Hire Atanas

Atanas Pavlov

Verified Expert  in Engineering

Software Developer

Gabrovo, Bulgaria
Toptal Member Since
October 21, 2015

Atanas has over ten years of experience working in software development, computer modeling, and data analysis. He led a joint industry project to develop computational tools for hydrogen fuel storage design. Through his company, he develops desktop and web-based computational software. Atanas has a passion for complex problems and new technologies for solving them. He truly believes in open source.



Preferred Environment

R, Scala, Clojure

The most amazing... I've developed is SmoWeb, an online computational platform for students, scientists, and engineers.

Work Experience

Senior Full-stack Developer

2014 - PRESENT
Precision For Medicine (via Toptal)
  • Created a significant part of the architecture of the web computational and visualization framework.
  • Developed much of the JVM back end of the framework and some of the front end.
  • Helped set up various projects using the framework.
Technologies: R, Scala, ClojureScript, Clojure

Software Architect | Developer

2010 - PRESENT
SysMo, Ltd.
  • Designed SmoWeb (a computational web platform) and developed its back end using Django and various Python computational libraries.
  • Designed SmoSTEM (a web application for scientific calculations) and developed its back-end using Flask.
  • Designed and co-developed SmoFlow, a C++ library for modeling fluid flow with heat transfer.
Technologies: Flask, Django, C++, Cython, NumPy, Python

Hydrogen Storage Modeler

2010 - PRESENT
BMW AG (via SysMo, Ltd.)
  • Developed computer models of hydrogen storage systems and simulated storage operations under driving conditions.
  • Built computational models of hydrogen fueling stations using liquid and gaseous hydrogen. Performed simulations of SAE J2601 compatible fueling stations.
  • Constructed a C++ system-level model for hardware-in-the-loop simulations of the hydrogen system controllers.
  • Developed a Python tool for computing material damage due to oscillating loads.
  • Created multiple Python tools for data processing including pre- and post-processing simulation results, and experimental results analysis.
Technologies: Modelica, Simulink, Django, Pandas, NumPy, Python, C++

Simulation Engineer

2006 - 2010
  • Developed a project plan for HySim (Hydrogen Simulation Tools) in cooperation with industry and university partners from ANSYS, LMS, and the Technical University of Dresden.
  • Implemented a library of components for system-level simulations (in C).
  • Developed system-level models of various automotive hydrogen storage system concepts (including 30 MPa cryogenic compressed hydrogen storage and 70 MPa compressed gaseous storage).
  • Developed and tested control concepts for automotive hydrogen storage systems.
  • Created concepts and computational models for hydrogen fueling stations.
Technologies: Simcenter Amesim, C, Simulink

A base numerical model class I developed, which was used for creating numerical models. These models were Python classes, which fully define a single-page web application, performing user input, computation, and result output.

SmoWeb: An Online Platform for Modeling and Computation
SmoWeb provides an infrastructure for the quick and easy development of one-page scientific web applications. The developer has to create a single Python class (a numerical model), define the input and output fields, and develop computational logic.

Based on this information, the platform automatically creates a web page allowing the user to input data, run computations, and inspect results. On our website, we have posted several applications from the fields of thermodynamics, heat, and mass transfer, and bio-reactor modeling, which we actively use in our work.

SmoSTEM: A Simple Tool for Engineering Modeling

SmoSTEM is a web application that allows users to create and share computational models. Boards with input and result fields are created (using drag-n-drop actions), and a subset of the Python language can be used for implementing mathematical operations (conditionals, loops, work with tables, and arrays).

QuartzBio | Interactive Web Platform for Biomedical Data Exploration and Analysis
Atanas had a leading role in architecting and developing a cloud framework for visualizing and exploring data from genomics, immunomics, and other biomedical studies. The platform presents raw data and results from various analytical pipelines in the forms of highly interactive tables and plots and allows the user to generate reports. Additionally, it provides an interface to some analytical and post-processing tools.

The platform was developed using Clojure, Scala, and R services on the back end and Reagent, React, and the ClojureScript framework on the front end.


Domain-specific Languages (DSLs), Clojure, Python, JavaScript, Scala, R, Simulink, Rust, C, Java, C++, Modelica, Cypher, Lisp


SciPy, NumPy, TensorFlow, Highcharts, Matplotlib, Keras, React, Node.js, Pandas, OpenCV, Flask-RESTful


RStudio Shiny, ClojureScript, Apache Spark, Flask, Django


Eclipse IDE, MATLAB, PyDev, Simcenter Amesim, Reagent, Aspera, Make, CMake, Git


MapReduce, Asynchronous Programming, Functional Programming, REST, Data Science

Industry Expertise



MySQL, SQLite, Amazon S3 (AWS S3), MongoDB


Control Systems, Data Modeling, Linear Algebra, Linear Regression, Machine Learning, Deep Neural Networks, Convolutional Neural Networks, Recurrent Neural Networks (RNN), Numerical Modeling, Cython, Digital Filters, Datomic, GraphDB, Algebra, Statistical Data Analysis


RStudio, Amazon EC2, Arduino, Linux

2004 - 2006

Master of Science Degree in Mechanical Engineering

Massachusetts Institute of Technology - Cambridge, MA, USA

2000 - 2004

Bachelor of Science Degree in Physics

Massachusetts Institute of Technology - Cambridge, MA, USA

2002 - 2003

Cambridge-MIT Exchange as Part of My B.Sc. Degree in Physics

Cambridge University - Cambridge, UK


Deep Learning Specialization