Gartner Critical Capabilities

Question 1

What is data science

Answer 1

Core data science is a practice centered on providing insight-driven decision making for improving business outcomes. Data and analytics leaders should use this set of critical capabilities for data science and machine learning platforms to govern, scale and maximize this practice.

Question 2

Key Findings

Answer 2

Data science and machine learning (DSML) platforms continue to democratize the usage of DSML techniques beyond purely code-driven development through GenAI- driven assistive features for data wrangling and automated insight generation.
■ Expert data science teams also benefit from GenAI-based code assistance for model development in notebook environments and the ability to deliver interactive chat- based applications for end users.
■ The need for collaboration between multidisciplinary technical and business teams is being addressed through workflow management, conversion between low-code and code-based models, and documentation generation.

Question 3

Recommendations

Answer 3

Redesign DSML operating models to take into account the augmented features of DSML platforms to decentralize more development work. Prioritize strategic initiatives for centralized expert data science teams.
■ Measure the productivity gains from introducing GenAI-based code assistants for notebook and code interfaces. Expand usage for the groups of users that derive the most benefit.
■ Expand lines of communication between disparate technical and business teams by developing and utilizing the governance and workflow capabilities of DSML platforms.

Question 4

Strategic Planning Assumption

Answer 4

By 2025, 90% of current analytics content consumers will begin creating analytics content enabled by AI.

Question 5

What You Need to Know

Answer 5

Analytics and AI leaders should utilize DSML platforms to advance the state of insight- driven decision making across business lines. This can be done by empowering expert data science teams with the tooling required to deliver end-to-end DSML projects, including critical phases such as design and planning, data preparation and exploration, model building and training, and ultimately data consumption.

Question 6

DSML Must-Have Capabilities

Answer 6

The must-have capabilities for this market include:
■ Import data from databases, data warehouses and file stores located on-premises and in the cloud
■ Build and evaluate models using a library of core data science and machine learning techniques, methods, algorithms and processes
■ Deploy, host and serve models in the platform for usage in services and applications

Question 7

Standard Capabilities

Answer 7

The standard capabilities for this market include:
■ Ability to build models from structured and unstructured data sources including text, images, video, audio and geospatial
■ Low-code interface for model development suitable for nonexpert data science roles, including business users and domain experts
■ Notebook-based code interface for data scientists to perform data access, preparation, model development and publication tasks
■ Postdeployment model life cycle management to retrain, retire or adapt models based on detecting and analyzing data, feature and model drift
■ Support for MLOps-based processes and tools that enable ML models to be deployed at scale in different operational environmentsThe standard capabilities for this market include:
■ Ability to build models from structured and unstructured data sources including text, images, video, audio and geospatial
■ Low-code interface for model development suitable for nonexpert data science roles, including business users and domain experts
■ Notebook-based code interface for data scientists to perform data access, preparation, model development and publication tasks
■ Postdeployment model life cycle management to retrain, retire or adapt models based on detecting and analyzing data, feature and model drift
■ Support for MLOps-based processes and tools that enable ML models to be deployed at scale in different operational environments

Question 8

Optional Capabilities

Answer 8

The optional capabilities for this market include:
■ Platform-generated recommendations for the best way to prepare, integrate and model data as well as automated creation of machine learning models based on manually selected target prediction
■ Advanced interfaces that facilitate more complex modeling for simulation, optimization and deep learning-based use cases
■ Custom SDKs that provide more control and flexibility for code-based model development and integration with services and applications
Functionality for working with GenAI models, such as large language models, through tracking, selection and monitoring of prompts, models and outputs
■ Techniques and tools that increase the transparency and interpretability of models to understand how and why model outputs are generated