Alaeddine Abdessalem


I am a Software Engineer, experienced in Backend Development, Data Engineering and Cloud Computing
Certified AWS Developer Associate

Technologies and Skills

Experience in building containerized and serverless web applications, designing and implementing data pipelines and building and deploying ML models

  • Backend development with Python, Django, Flask, Docker, AWS and GCP
  • Programming Languages : Python, Java, Javascript, PHP, C/C++, C#
  • Databases : PostgreSQL, MySQL and DynamoDB
  • Big Data : Hadoop, Kafka, Spark, ElasticSearch and Logstash
  • Machine Learning : Keras and Tensorflow
  • NLP : NLTK and StanfordCoreNLP
  • Web Scraping with Selenium and PhantomJS
  • Amazon Web Services (AWS) Certified Developer - Associate

Career



Backend Intern at Myfi Networks, Sakarya, Turkey- Summer 2019

During this summer, I had the opportunity to work in Myfi Networks in Sakarya, Turkey, as a backend engineering intern.

My mission was to create a network automation and monitoring platform. The tech stack was Python and Flask for developing the Restful API and backend scripts, Postgres as the data store, PHP, Socket.io, Elasticsearch and Logstash.




Backend and Data Engineer, Scout And Cellar- 2019

I worked as a freelancer for Scout And Cellar and I was in the Backend team, where I participated in designing the data pipelines and ETL and building their microservices.

Projects

MaintainIt

This project was developed during the OpenGovDataHack2020 hackathon and got us the first prize. The project aims to reduce flight delays caused by technical issues by predicting the Remaining Useful Life (RUL) of components in the aircraft.

During 48 hours, we trained a Gradient Boosting Regressor model on the Turbofan engine dataset to predict the RUL from sensors data, had it deployed,integrated it in a webservice and exposed our solution as a dashboard to monitor the health of the aircrafts.

You can check project's source code in 2 versions with Django and Flask, here and here. It's also deployed and accessible in this link :

EIDR Data Mining- ETL

The project consists in creating a relational database for all EIDR content and pulling that content using web scraping, text processing techniques and MySQL queries. Multithreading was also a very important technique to accelerate web scraping process.

This project provided a well structured database that was useful for my client's business related to all audio-visual content produced around the world.

PFNN

Phase-functioned neural-networks is a new deep learning technique which is very efficient to model human and animal mouvement. It produces realistic animations for characters for complex terrains. Using the public research paper, I implemented the algorithm to work reliably on Unity 3D and I delivered the project for my client to be used for game industry.

Skills involved : deep learning, tensorflow, GPU computing, openGL, GPU programming, Kinematics...

Keywords extraction algorithm

In order to automate feedback analysis for my client's e-commerce business (targeted for chinese communities), I wrote a python script which extracts all important keywords from chinese text with their importance scores and sentiment values. In this project skills like Natural Language Processing and deep learning were essential.

Technologies/ Libraries used : Stanford CoreNLP, Keras, Regex

Quadcopter

For my high school graduation project, I created a quadcopter. In this project, I learned skills like MCU programming, schemtics design, PCB design, CAD, wireless communication (Zigbee, NRF, WIFI), PID control, flight controller algorithms, gyroscope/accelerometer use and data filtering and processing. We designed many versions with different frames, communication modules and control algorithms.

QuadTree implementation- Crashy.io

My client was running a naive O(n²) collusion detection algorithm implemented in his game's server node.js program. The algorithm was too slow that the server couldn't hold many players at the same time. I implemented the O(n log(n)) QuadTree-based collusion detection algorithm which made the program run much faster and enabled the server to hold much more players.

QLearning agent

I implemented the Qlearning algorithm (which is a reinforcement learning technique) to make an intelligent agent solve a maze in Unity 3D's environment.

Prizes and Achievements



First Prize at OpenGovDataHack 2020

OpenGovDataHack2020 is the first national hackathon promoting open data, in Tunisia. The hackathon aims to encourage the use of open data by developers and entrepreneurs to improve public sectors. During this hackathon, we worked on reducing the flight delays of Tunisair (a national airline), by developing a dashboard to monitor the health of aircrafts. Getting the first place required us to train a predictive model, deploy it on the cloud (AWS) and integrate it in a dashboard to help the administration avoid unexpected component failures.

Tech Stack : Python, Flask, AWS EC2, PostgreSQL, Keras, Gradient Boosting





Maghreb Champions during the ACPC 2019 Arab and Africa Collegiate Programming Championship

We were crowned as the Maghreb Champions during the ACPC Arab and Africa Collegiate Programming Championship. ACPC is a competitive programming contest that gathers teams from arab and african countries, who qualified from national contests. ACPC was a good opportunity for us to demonstrate our problem solving, algorithmics and data structures skills.





First prize at NIT day competitive programming contest 2018

This contest that was held during the NIT day 2018, is a 5-hour long challenge where contestants solve problems involving algorithmics and data structures and they are required to write efficient programs. This contest allowed me and my team to demonstrate our skills in problem solving and programming.





Guarantees

Professional Development

One of my important concerns is to write scalable, maintainable, well documented and efficient code. I always make sure to respect programming conventions and to use the right tools, design patterns, data structures and algorithms to provide the best solution.

Project Analysis and Design

I consider project analysis and design a crucial step in every project. Designing the right diagrams for the system according to industry standards and choosing the most suitable tools, design patterns, frameworks and infrastructure not only optimizes time and efforts but also reduces the project's budget and ensures having a solid and scalable system.

On Time Delivery

I make sure to deliver projects on time, respecting my client's requirements and deadlines.

What clients think

I am also a freelancer on Upwork

Contact me for more details.

See Upwork Profile