Written by Vadym Bartko

Vadym Bartko - Python Backend Engineer

My CV in PDF

Contacts

COMPUTER SKILLS AND COMPETENCES

AWS: Lambda, SQS, SNS, S3, Textract, RDS, DynamoDB, MemoryDB

Backend: Python Backend Stack - Flask, Django, DRF, gunicorn, RabbitMQ, Celery, boto3, Tornado, Flasgger, dynaconf, MongoDB, ElasticSearch etc

Databases Postgres, RDS, MemoryDB(Redis), DynamoDB

Pdf Processing: ghostscript, fitz(mupdf/pymupdf), deep analysis of pdf documents, font related issues, text detection

AI: Convolutional Neural Networks, image classification, object detection, image captioning, image segmentation(UNet), homography, keypoints detection, image processing, FaceNet, classification and clusterization methods, ROS, datasets preparation

AI Tools: OpenCV, Scikit-learn, Scikit-image, Tensorflow, Pandas, Jupyter Notebook, Keras, Dlib, Pillow

Languages: Python, Java

OS: Linux (Debian based), Windows

Cloud and Deployment: Kubernetes, Docker, nvidia-docker, Helm, AWS, microservices and distributed systems

Other: some experience with LIDARs and stereocameras, R&D, basic language processing

NOTABLE PROJECTS

[miltech]

radio scanner based on SDR and single board computer (orange pi), reactive jammer (counter drone), hardware and software design

[pdf processing - fintech]

processing and analysis of financial pdf documents, OCR, dealing with pdf format issues, pdf/image compression and optimization, ghoscript based processing, dealing with font related issues

ghostscript, imagemagic, fitz, flask, AWS Lambda, AWS Textract

Khumbu

image search engine based on AI and computer vision. Performs face recognition, object detection, image classification, metadata extraction, query processing.

Docker, RabbitMQ, Flask, Mongo, Keras, Tensorflow, Kubernetes, React

technical research, market research and analysis, prepare pitch deck, interviewing potential customers, talk to investors, implement backend and frontend parts

[bots and image processing]

automatically train customer specific NLP bots, deploy ML apps to cloud Kubernetes, Helm, nvidia-docker, Flask, Celery, Airflow

implement web api for third-party ML models, build distributed training system, Kubernetes management and deployment, performance optimization

[media assets management software]

face detection and recognition, image tagging, inappropriate content detection

OpenCV, Keras, Tensorflow, Docker, dlib, pandas, scikit-image

implement image classification and nsfw content detection, build dataset for face recognition tasks, test various facerecog approaches, integrate with another services

[autonomous robocar]

control, navigation, route planning and map building for small 4W robot OpenCV, ROS, LIDAR, ZED Camera

hardware design, test various SLAM algorithms, setup ROS env for wheel robot, experiments with LIDAR and stereocamera

[gene variations classification]

detect pathogenic variations in genes pandas, scikit-learn

ITraffic

detect empty parking lots using webcamera Opencv, Caffe, digits, Python, Docker, RabbitMQ; AKAZE/ORB/BRISK, homography

create datasets, train models, implement API for models, deployment, implement cv tasks

[crawling/scraping]

large scale web crawling and data extraction

RabbitMQ, MySQL, Python, PhantomJS, RedShift, S3, EC2, ElasticSearch, NLTK

Firefly

web manager for scientific papers

GridFS, TitanDB, ElasticSearch

Nomad

distributed crawler

TitanDB, OrientDB, gremlin

Denigma

tools for ageing research project, server administration, knowledge extraction

CloudStack, Scala, NLTK, OpenNLP, semantic tools

research, backend implementation, some DevOps

BigPlanet

offline map with different sources of geo data

Android 2.x

WORK EXPERIENCE

EDUCATION AND TRAINING

OTHER