Personalized search; Recommendations; Feedback learning
Shaped provides relevance infrastructure for search, recommendations and agent context. It connects data sources, manages embeddings and uses interaction feedback to personalize retrieved results. A query can include text, a user identifier or an item identifier, allowing the system to use the available context. The official site also describes hybrid search that combines semantic and keyword retrieval with behavior-based ranking. Python and TypeScript SDKs and MCP access are presented as integration routes. The homepage states that Shaped has been acquired by Whatnot, an important current-status detail. Product availability and commercial arrangements should be confirmed with the company rather than inferred from an earlier standalone offering.
Shaped is best described as Search And Knowledge Tool for researchers, knowledge workers. The practical workflow centers on personalized retrieval, hybrid search, interaction-feedback ranking, recommendations, developer sdks. Users normally bring their own task context into the product and review the resulting output before relying on it.
Useful use cases include personalized search, recommendations, feedback learning. Category placement is kept to Search And Knowledge because the tool should be listed where people would actually compare it. Supported access is recorded as the access model described by the product, and integrations are limited to connections that could be verified from the available source material.
The developer is recorded as Shaped. Pricing is listed conservatively as Current pricing should be checked on the official site. Free-plan status is recorded as Free access terms are not clearly established in the reviewed public material. Before using Shaped for production work, check the current plan page, account limits and any commercial-use terms that apply to the files, data, media or decisions involved.
Run a small real task first and compare the result with the original material. For generated text, media, code, analysis or operational actions, review factual claims, permissions and handoff steps before publishing or applying the output. This keeps the listing useful without adding unsupported benchmarks, invented model names or broad legal promises.
