Tech Stack
DockerA/B TestingGoMLflowAWSAirflowC++PyTorchunittest
Job Description, Responsibilities & Requirements
Description language:
About the Position
Impressit is looking for an MLOps Software Engineer to join a cross-functional team building the infrastructure behind autonomous AI trading systems, as part of a larger trading division. This is a great opportunity to contribute to a meaningful product with real-world impact, while working with modern technologies and international teams.
Responsibilities
- Own MLOps end-to-end - from model training and evaluation to trading evaluation
- Develop large-scale deployment of GPU nodes running across dozens of Kubernetes clusters across regions, using CUDA, Python, PyTorch, Triton, Redis, NCCL, NVLink
- Build simulation pipelines to scale hyperparameter testing for model training and trading, synchronizing model parameters with financial trading parameters
- Automate feature evaluation and experimentation for model training
- Automate model training and performance evaluation
- Combine classical model evaluation with financial trading metrics - Sharpe, Turnover, Fitness, Returns, Drawdown, Margin, Long/Short Count, Sector/Subindustry Allocations, etc.
- Build, maintain, and enhance model pipelines using MLOps and AIOps frameworks
- Work across the full data and model infrastructure stack, including SageMaker Studio
- Partner with Data Scientists to set up model monitoring and feedback systems
- Be part of a sub-team focused on fully autonomous AI trading within a larger trading division
Requirements
- Bachelor's degree or above; a Software Engineering background is preferred
- Hands-on experience with CUDA and PyTorch
- Strong knowledge of low-level languages such as C++ or Go
- Experience with at least one of the following: high-scale distributed training and inference/ML systems; parallel and distributed computing with multiple GPUs; high-throughput scheduling as a service at supercomputing scale; or the ability to resolve Segmentation Faults
- Proven experience with Unit Tests, Integration Tests, and error handling to keep trading systems fault-tolerant and robust
- Experience with orchestration tools - DAGs in Airflow or Prefect, MLflow
- Experience with deployment - Docker, AWS, load testing
- Good communicator, comfortable working in cross-functional teams
Nice to Have
- Broad, in-depth understanding of network technologies
- Prior experience in fintech / algorithmic trading
- Passion for clean code, developer experience, and end-user quality
We Offer
- Professional education & training budget
- 24 working days of paid vacation
- WOW team-building events
- Paid sick leave
About the Company
We are passionate about everything we do - and we back the people who do it.
Ready to join?
Send your CV - it takes two minutes, and a real person reads every application.