Want to connect with Customiser?
Join organizations building the agentic web. Get introductions, share updates, and shape the future of .agent.
Is this your company?
Claim this profile to update your info, add products, and connect with the community.
Customiser is a vertical AI company that uses a multi-agent orchestration pattern to solve complex data extraction problems. Unlike simple chatbots, their platform employs a pipeline of specialized agents—Classifier, Extraction, Cross-Reference, and QA—that work in sequence to perform high-stakes analysis. This makes them a relevant example of how autonomous agents can be applied to specific, high-value industrial workflows where accuracy and verification are more important than general-purpose conversation.
In the broader agent ecosystem, Customiser sits at the intersection of Document Intelligence and Agentic Workflows. They are particularly relevant for their use of "Knowledge Bases" to ground agent behavior in proprietary business standards. This demonstrates a clear path for enterprise agent adoption: provide a framework where agents are not just processing text, but are cross-referencing and validating findings against a user-defined ground truth.
Customiser is a platform for automating the extraction and analysis of technical data within industrial sectors such as manufacturing, construction, and oil and gas. While generic AI tools often struggle with the dense, non-standardized formats of industrial documentation, Customiser uses a modular architecture of specialized agents to process these files. The company was founded in 2024 by a team including CEO Michael Adomako, COO Saban Mehic, CTO Betim Sherifi, and CPO Fron Hasani. They operate as an independent workspace focused on bringing order to the scattered information found on production lines and factory floors.
Industrial operations are typically buried under a mountain of PDFs, spreadsheets, and technical drawings. The challenge for these teams is not just reading the text, but verifying that the data in a drawing matches the Bill of Materials (BOM) and complies with customer specification codes. Customiser addresses this by allowing users to build "Knowledge Bases"—structured databases of internal standards, supplier directories, and pricing catalogs. These serve as the ground truth for the agents during analysis.
The core of the platform is a sequential pipeline of specialized AI agents. When a document set is uploaded, the Classifier Agent first identifies the document types and routes them to the appropriate specialists. Extraction agents then pull data according to user-defined schemas. Unlike many competitors that use fixed templates, Customiser users write their own extraction prompts and define JSON output formats, providing more control over how data is structured.
Perhaps the most distinct feature of the platform is the Cross-Reference Agent. This agent compares extracted data points across different documents, identifying matches, mismatches, and gaps. For example, it can automatically flag if a finish requirement in a customer spec does not appear in the final submittal. A final QA Agent reviews the work of the previous steps to generate a summary and highlight critical findings for human review. This multi-step process aims to reduce manual data entry and catch errors before they lead to costly production mistakes.
Customiser is built with an enterprise-first approach to security and deployment. Recognizing that technical drawings and proprietary specs are highly sensitive, they offer data residency controls, role-based access, and Zero Data Retention (ZDR) options. The platform uses a credit-based pricing model where users only pay for document analysis, while setting up schemas and building internal knowledge bases is free.
By moving the document processing workflow from simple OCR to agentic reasoning, Customiser enables faster decision-making for operations teams. The founders claim significant efficiency gains—up to 85% in some cases—by replacing manual review with structured, searchable intelligence. As manufacturing systems become increasingly complex, Customiser provides a way to unify scattered data into a single source of truth that agents can actually use.
Configurable AI agents that extract structured data from technical documents and cross-reference specs against BOMs.
Internet Computer blockchain source: the client/replica software run by nodes
Rust bindings for Binaryen's wasm-opt
Ethereum JSON-RPC multi-transport client. Rust implementation of web3 library.
Standards for the Cosmos network & interchain ecosystem.
Submit extrinsics (transactions) to a substrate node via RPC
Customiser is hiring
You've explored Customiser.
Join organizations building the agentic web.