FAQ from SuperDuperDB
What is SuperDuperDB?
SuperDuperDB enables seamless AI integration within your existing database, allowing real-time inference, model training, and vector search—all executed through Python without data migration.
How to use SuperDuperDB?
Connect your database, define your AI models in Python, and let SuperDuperDB synchronize everything. Predictions happen automatically during queries, and training is triggered via standard data access patterns.
What can I do with SuperDuperDB?
You can deploy AI models for automatic inference, train models directly on stored data, enable vector-based semantic search, and integrate third-party AI services—all inside your current data infrastructure.
Which databases are supported by SuperDuperDB?
Supported databases include MongoDB (Atlas), Amazon S3, PostgreSQL, MySQL, DuckDB, SQLite, Google BigQuery, and Snowflake.
Can I train and fine-tune models using SuperDuperDB?
Absolutely. Train or adapt models—including deep learning and classical ML—by simply querying your data. SuperDuperDB handles feature extraction and incremental learning behind the scenes.
Can I integrate AI APIs with SuperDuperDB?
Yes. You can plug in APIs like OpenAI, Anthropic, or Cohere and combine their outputs with locally hosted models for hybrid AI workflows.
What ML/AI frameworks are supported by SuperDuperDB?
Native support includes PyTorch, TensorFlow, Scikit-learn, HuggingFace Transformers, Keras, and more—enabling flexibility across the AI stack.
Is SuperDuperDB suitable for Full Stack Developers?
Definitely. Full stack engineers can embed AI features into apps quickly using familiar tools, without needing MLOps expertise or managing extra services.
Is SuperDuperDB suitable for Data Scientists?
Yes. Data scientists retain full control over modeling while reducing overhead—use your favorite libraries and iterate faster with direct access to production data.
Is SuperDuperDB suitable for ML Engineers?
Yes. ML engineers benefit from a unified, scalable platform that supports deployment across local, on-premises, and cloud environments with consistent tooling.