Crisp Launches Halo to Bring Retail AI to Enterprise Cloud Environments – Unite.AI

Crisp launched Halo, an AI platform for retail that can run within a company’s cloud environment, bringing sales and supply chain analytics closer to the data and governance systems companies already use. Announced on October 29, the platform is aimed at manufacturers, retailers, distributors and intermediaries of consumer packaged goods (CPG).

The launch addresses a familiar retail problem: Teams may see a change in sales or inventory, but they still spend hours piecing together the context needed to explain it. Crisp’s Halo announcement describes a system that combines retail data, industry-specific logic, and agent tools to help users investigate performance and decide what to do next.

A database for retail under artificial intelligence

Halo’s architecture begins with information coming in from resellers and distributors. The Crisp platform overview separates that foundation from data integration, management and modeling, with agent tools and role-specific applications built on top of it.

This structure is important because a product, store or reporting period can be represented differently between business partners. A question about an inventory shortage becomes more useful when the underlying systems agree on which item, location, and time period the question is about.

Crisp describes his approach as vertical AI: data is cleaned and enriched with retail semantics before being used for analytics. The company’s background materials identify retailer calendars, product hierarchies, fill rates, and out-of-stock analysis as examples of the business context involved. Crisp says more than 7,000 customers, including 90 of the top 100 CPG companies, use its services; this is the adoption data reported by the company.

Run inside the customer cloud

Halo is available as a hosted platform or within customer environments on Databricks, Snowflake, and Google Cloud. The latter approach offers companies an alternative to moving their retail intelligence to another externally hosted application.

Crisp’s Databricks architecture documentation provides a concrete example. It inserts retail data into Delta tables managed through Unity Catalog, hosts applications and agents in Databricks Apps, and uses a Lakebase database for application state, conversations, and agent storage. Model calls are performed via Model Serving and AI Gateway, while semantic queries use a selected SQL warehouse.

According to Crisp, proprietary data remains in the customer’s environment, while non-proprietary operational logs and telemetry are shared with its central systems. This distinction is more precise than suggesting that deployment has no external communication. Crisp also lists the integration of MLflow and the publishing of its semantic models as metric views of the Unity catalog as roadmap items.

For IT teams, the importance is operational: access controls, inference choices and usage monitoring can align with existing infrastructure. Deployment location alone, however, does not establish the accuracy of a response or the suitability of an automated action.

From wallet questions to retailer-specific decisions

The launch introduces Central Desk and Customer Desk. Central Desk gives head office teams insight into their retail portfolio for work spanning sales strategy, category management and supply chain planning. Customer Desk is aimed at account teams that need the metrics, terminology and operational logic of a single reseller.

Consider a brand investigating weaker sales for a particular product. The useful question is whether the change reflects demand, availability, distribution, or another operational factor. Linking analytics to store- and item-level records can help teams narrow their investigation rather than treating an entire portfolio decline as a single issue.

This is the practical distinction between a conversational interface and a retail application: the latter must understand the business entities and relationships behind the application. Halo’s broader platform also organizes capabilities around functions such as replenishment, category planning and supplier collaboration.

Trackable responses and repeatable workflows

Crisp’s AI agent materials describe tools that use everyday store- and product-level intelligence to support analysis and action. In Halo, the company lists conversational questions, recurring reports, workflow automation, and custom applications as the primary ways to use those tools.

The launch materials emphasize tracking recommendations on the source data and the logic used to produce it. Such provenance can make the answer easier to verify: an analyst can examine the documents and reasoning behind a recommendation instead of accepting a fluent explanation at face value.

For companies evaluating the platform, the important questions range from explanation to execution: which actions are allowed, which require review, and how changes are measured after implementation. Published materials describe capabilities and implementation architecture, rather than an independent benchmark of forecast accuracy or financial returns.

Halo’s strongest proposition is the connection between these layers. The retail context gives an AI system more useful information to work with, the customer cloud implementation can adapt to existing governance, and role-specific workflows put the resulting analytics in front of the people responsible for inventory and business decisions.