Block-based workflow editor
Prompting, testing, versioning, multi-prompt, cost control
Resume line
Retrieval Augmented Generation (RAG)
Roles and permissions
Security, data transparency
Web
Data Transformer
Decision Maker
Extract Maker (with HTML)
Key / Knowledge Manager
Web Creator
SaaS Email, Integration
Cloud privacy managed
On premise
CASE STUDIES
AI in action in the real world
Discover how different companies, from the food industry to banking, are using ARI to transform their processes, improve efficiency and create new value.
Company operating in the food industry
Optimizing Order Management with Artificial Intelligence
The Challenge
The company received orders via email from various clients, with PDF attachments generated by different CRM software. Each client used different units of measurement and ordering methods: pieces, cartons, kilos, grams, and so on. Some locations ordered directly, while others centralized orders through a hub.
The Solution with ARI
We implemented an artificial intelligence flow with ARI that hooks into incoming emails, distinguishes orders from other communications and interprets attached PDFs. ARI understands each customer's specifications and translates the data into the correct format for the company's management system.
The Review Process
Once interpreted by ARI, orders pass through a dedicated backend interface. This interface shows the human operator what operations were done, compares new orders with database history and flags which are perfectly in line with past orders, which have minimal differences and which are dissonant. This way, the operator can manually verify and approve data before orders are finalized.
The Result
Thanks to this system, the company has obtained an automated, transparent and reliable order management process, reducing errors and ensuring human control where necessary.
CASE #01
Bank specializing in pawn loans
Supporting Asset Valuation in a Pawn Bank with Artificial Intelligence
The Challenge
The bank wanted a system that could provide sophisticated support to human valuers, offering more precise value estimates based on current and historical data. It was crucial to maintain the confidentiality of the methodologies used, as these are part of the bank's exclusive know-how.
The Solution with Complex Concurrent Flows
We created a series of parallel flows that compete to provide an extremely accurate valuation. For confidentiality reasons, we cannot go into detail about each flow, but we can give a significant example. For example, one of these flows uses photographs of objects to estimate precious material content. In practice, the system identifies gold within a watch by cross-referencing photographic information with dimensions and a reference database, to calculate the actual weight of gold present. Similarly, other flows work in parallel to consider market value and other parameters.
The Result
Thanks to these integrated flows, the bank can offer its valuers a level of accuracy and precision that goes beyond simple human valuation. The result is a more reliable and competitive valuation service, while still maintaining professional secrecy about the methodologies used.
CASE #02
Insurance broker
A Smart Expert for a Banking Services Broker
The Challenge
The broker needed a system that would allow banks to offer their customers clear and precise information about selected additional services. It was crucial that the system be limited to bank-approved services, avoiding providing off-topic or inaccurate information, as the end customers are the banks' customers.
The Implemented Solution
We developed a Smart Expert in the form of a specialized chatbot. This chatbot, integrated into the broker's platform, responds exclusively about these additional banking services, customized based on the bank's choices and established parameters. In this way, the chatbot behaves like a sector expert who stays within the boundaries of authorized information, ensuring that bank customers receive only accurate and relevant details.
The Result
With this system, the broker was able to offer banks a reliable and precise tool. Banks, in turn, can offer their customers additional services without risking providing incorrect or irrelevant information, improving their offering in a safe and controlled way.
CASE #03
HR Software
Enhancing HR Software with Artificial Intelligence
The Challenge
The HR software needed to evolve to more efficiently manage internal processes, automate some functions and provide intelligent support for financial and personnel management decisions.
Implemented Features
We integrated several AI-based features. One of the main ones concerns expense reimbursement management: starting from a simple photo of a receipt or invoice, the system extracts the data and compares it with company policies. Automatically, it provides the finance department with an indication of expense compliance and approvability. It's also possible to request human review through a control panel, leaving the final decision to the financial manager. Another implemented functionality is automatic document insertion: both personal documents like ID cards and certifications and other company documentation are scanned and automatically inserted into systems, simplifying administrative management.
Future Features
In the future, the software will also include advanced CV management, capable of processing non-standardized resumes and comparing top candidates, verifying web information and validating data. Additionally, the introduction of an onboarding chatbot for new employees is planned, which will answer common HR questions and help new hires get oriented.
CASE #04
Banking App
Automating Tax Deduction Transfers in a Banking App
The Challenge
The bank wanted to offer its customers a tool that would simplify the payment process and make it easier to correctly obtain tax deductions. The goal was to recognize quote details simply by taking a photo and then generate a tax-compliant transfer.
The Implemented Solution
We created a feature that, starting from a photograph of a quote, reads and interprets the items present. Based on the type of service or product described, the system recognizes which tax deductions are applicable. Consequently, it automatically fills out a transfer that meets the requirements necessary for the deduction, making the process simple and secure for the user.
The Result
With this feature, the bank's customers can transform a quote into a tax-deductible transfer with a single shot. This not only speeds up operations but also reduces errors and ensures that all paperwork complies with current tax regulations.
CASE #05
Newspaper
Creating Internal Link Building for a Newspaper through ARI and a RAG System
The Challenge
The newspaper wanted to build robust internal link building, allowing journalists to intelligently link new articles with past ones. The goal was to enrich the narrative and keep readers engaged, avoiding repeatedly rewriting the same information and creating coherent internal links.
The Implemented Solution
Thanks to ARI, we developed a RAG system that analyzes the content of a new article and suggests links to previous articles in real time. For example, if a journalist writes a piece about a thwarted bombing in Ukraine thanks to Patriot missiles, the system can suggest a link to an old article that covered the supply of those missiles to Ukraine.
The Result
In this way, the newspaper creates a network of internal links, improving historical coherence and enriching the reader's experience. The RAG system based on ARI integrates directly with the editorial workflow, helping journalists build solid internal link building and keep the narrative alive over time.
CASE #06
Data Whisper
Data Whisper - The AI Assistant for Managers
The Challenge
Many companies have large amounts of data scattered across structured and unstructured databases. Managers often have difficulty querying this data quickly and getting answers without knowing query languages. Additionally, we wanted to offer a tool that could trigger alerts and support meetings by providing contextual data in real time.
The Implemented Solution
"Data Whisper" was designed as an AI assistant for managers. It has three main features: 1. Natural Language Querying: Managers can ask questions to their databases in natural language. 2. Custom Alerts: The system can set specific alerts to proactively inform about important situations. 3. Meeting Support: Data Whisper can listen to discussions and automatically present relevant data to the context.
The Result
"Data Whisper" represents an innovative feasibility study. With its ability to make data accessible, send alerts and actively support meetings, this tool can transform how managers interact with company data, making every decision more informed and timely.
CASE #07
PLATFORM
request a quote
Create, orchestrate and publish GenAI solutions securely, quickly and under your total control.
Production Ready Interface
Inspired by Ariadne, mythological symbol of guidance through complexity, ARI is an all-in-one development platform with a clear mission: to make the creation of Gen AI solutions fast, intuitive, and accessible to everyone.
ARI is a product by BlackBytes SA Serafino Balestra 10, 6900 Lugano, CH CHE-189.438.083
Get in touch
[COOKIE_POLICY]
> SYSTEM NOTICE
We use cookies to improve your experience on our site You are free to decide whether to allow them or not.