cnvrg.io

Cnvrg.io helps AI developers by giving resources to create and deploy ML models.

Automate

• End-to-End: Helps to build production-ready ML models
• Portable: Makes difficult engineering tasks simple with the help MLOps and container infrastructure
• Optimized: Designs ML models to run on any AI stack to achieve maximum performance

Orchestrate

• Metacloud Utilization: Allows to access any cloud or industry-specific cloud in only one click
• Visibility & Control: It maximizes utilization of servers for all models that run with a 360-degree view of your whole AI infrastructure by project, workload, container and job.
• Dynamic Orchestration: Using native Kubernetes and meta scheduler it helps to mix and match cloud resources for different ML pipelines

Research

• Collaborate: Assists ML projects with interactive dynamic interfaces
• Track: Saves models and meta-data using tracklog, including parameters, code version, artifacts production models
• Analyze: You can compare and visualize the training and deployment metrics of your experiments in real-time.

Deploy

• Production Ready: You can deploy your models at high speed with autoscale and advanced build in monitoring capabilities.
• Compliant: Track inference on top of models to maintain performance and update the models in real-time.
• Delivery:  You can publish your models in just one click on Kubernetes clusters via web-service, RabbitMQ, Kafka stream etc.

Manage

• Easy access: You can access all your team’s data in one central hub for any project or task
• Support all data types: You can import data in any format and integrate it into any experiment or analysis session
• Machine learning ready: Manage datasets with tagging, automated versioning and querying capabilities.

Solutions

• Churn Prediction
• Image Classification
• Custom AutoML
• Text Analysis and Natural Language Understanding
• Recommendation Engine
• Reinforcement Learning
• Data Management
• Fraud Detection
• Speech Recognition

Industries

• Insurance
• Manufacturing
• Healthcare
• Financial Services
• Automotive
• Gaming
• Media
• Pharmaceutical
• Retail
• Aerospace and Defense

• Dataset Management
• ML Model Deployment
• Open Compute
• AI Library
• Machine Learning Tracking
• Machine Learning Pipelines
• Scalable Streaming Endpoints

1. Free 0$
• Infrastructure: 8CPUs and 16GB UNLIMITED USAGE
• Support: Community Platform
• SELF MANAGED: Fully managed service, BYO compute & storage or hybrid
• RESOURCE TYPE: CPU
2. Pay as you go
• Infrastructure: Can add additional resources as you need
• Support:24/7 customer service
• UNLIMITED RESOURCES MANAGEMENT: YES
• RESOURCE TYPE: UNLIMITED
3. Enterprise
• Infrastructure: Leverage volume discounts
• Support: 24/7 customer service
• CUSTOMER ACCESS: Dedicated customer success engineer
• UNLIMITED RESOURCES MANAGEMENT: YES
• RESOURCE TYPE: UNLIMITED
4. Self Hosted
• CUSTOMER ACCESS: Dedicated customer success engineer
• SELF MANAGED: BYO - Connect any resource on-prem or cloud
• UNLIMITED RESOURCES MANAGEMENT: YES
• RESOURCE TYPE: UNLIMITED