Take offline models from ML scientists and turn them into a real machine learning production system
Build and deploy training and serving pipelines for ML models in GCP
Identify and evaluate new technologies to improve performance, maintainability, and reliability of existing machine learning systems
Work with upstream and downstream engineering teams to define the data contract for serving in platforms such as KubeFlow and Vertex AI
Work as a liaison with ML Science and Data Engineering team to understand the data schema and convey changes and updates to the respective teams
Scope latency issues for multi-cloud deployments (AWS vs GCP) to enable real-time ML serving
Enable dataset permissions, onboard pip packages to enterprise ML platform registry for Security approval, creating custom Docker images from base Python images
Qualifications :
Experience developing with containers and Kubernetes in cloud computing environments (AWS/GCP)
Familiarity with one or more data-oriented workflow orchestration frameworks (KubeFlow/Airflow/ Argo, etc.)
Exposure to deep learning approaches and modeling frameworks (PyTorch/ Tensorflow/ Keras, etc.)
Strong experience in writing SQL queries for production systems (with BigQuery or other data warehouse tools)
Experience working as one or more of the following DevOps Engineer, SRE, Platform Engineer, Infrastructure Engineer, Cloud Engineer, and/or Production Engineer.
Familiarity and working knowledge with Kubeflow, Seldon Core, Artifact Registry
Knowledge of Unit testing in Python, Mocking, Pytest.
Knowledge of model serialization and deserialization for Pytorch, Scikitlearn and Tensorflow
Knowledge of Git and Github. Gitops, Bazel and CI/CD deployments with Jenkins
Understanding of MLOps, Model development lifecycle with knowledge of Training and Deployment pipelines for Machine Learning solutions.
Nice to have model monitoring and debugging experience in production
Notes
* Topcoder does not provide visa sponsorship nor will we work with Staffing Agencies.** USA Visa Holders - Please consult an attorney before applying to any Topcoder Gig. Some visa statuses will or will not allow you to conduct freelance work with Topcoder.*** Topcoder and Wipro employees are not eligible for Gig work opportunities. Do not apply and send questions to talent.topcoder@wipro.com.
At Topcoder, we pride ourselves in bringing our customers the very best candidates to help fill their needs. Want to improve your chances? You can do a few things:
Make sure your Topcoder profile says it all. Fill out your profile to the best of your ability. Your skills, your location, your devices, etc, all help you improve your chances of being selected for a gig.
Let us know you’re here! Check in on our Gig Work forum and tell us you’re looking for a gig. It’s great visibility for the Gig team.
Check out our Topcoder challenges and participate. Challenges showing your technology skills make you a “qualified” candidate so we know you’re good. The proof is in the pudding!