We are seeking an experienced platform engineer to to scale our continuously learning systems. The platform you will be developing is at the heart of model creation and deployment processes.
It allows clients and Blackshark.ai to manage annotated data, trigger long-running computations and store data streams from devices.
Over the years, we gathered what makes and breaks such a platform. Every iteration brought us closer to the optimal solution.
Nevertheless, there are many challenges left. Here, we need your expertise, creative problem solving skills, and desire to make an impact.
Our Tech Stack :
DL Frameworks : Pytorch, Torchvision, CUDA
Frontend : VueJS
Backend : Django, FastAPI, uvicorn, uwsgi
Data : PostgreSQL, PostGIS, RabbitMQ, SQLAlchemy, Logstash
Traditional Computer Vision : OpenCV, GDAL, scikit-image, pillow, Shapely, pyWavelet
Packaging : Docker, Kubernetes
Deployment : Gitlab CI / CD, Jenkins
Fluent in Python 3
Relational DBs, Data Modelling, SQL
Experience with Docker / Kubernetes deployments
Fundamental OS know-how, Unix and its command line tools
You easily pick up new technologies, and have a curious and inquisitive mind.
You are a good communicator, enjoy working in a team, and are quick on your feet.
It’s nice if you can show practical experience in any of those areas :
Machine Learning Systems in cloud-only or hybrid (online / offline) scenarios
Critical ETL workflows where data is large in-memory or multiple TB in cloud storage
Managing Cloud resources with infra-as-code approach
Workflow and Task Management Systems
Programming skills and knowledge that shows your passion for CS
Software engineering skills (testing, deployment, configuration management)