Proof, not promises.
The 5% of AI projects that reach production have one thing in common: they learn. Here is the work - across finance, construction, manufacturing and more - that crossed the GenAI Divide and shipped.
LLM-powered Quotations and Reporting in Construction
Read moreLLM-powered Quotations and Reporting in Construction
Our client wanted to reduce time spent on providing manual quotations for their customers in the construction sector.
We designed an LLM-based system that uses the past 10 years of historical data on customer accounts and quotations to automatically produce quotes for new clients.
Thanks to AI all new team members will have instant access to expert knowledge of construction engineers and architects in their company, increasing the oveall team intelligence.
LLM in Freight Investment
Read moreLLM in Freight Investment
Brainpool team of AI consultants designed a strategy to integrate the "Freight GPT" AI system into the Freight Trading Platform to improve accuracy and efficiency of its operations.
The AI model was designed to accurately price each shipment based on various data points available, such as shipping congestion, seasonality, cargo types.
AI assistant will disseminate information to users, with an initial focus on specific ports in the South China Sea and another undisclosed location.
LLM for Parking Ticket Appeal Process Automation
Read moreLLM for Parking Ticket Appeal Process Automation
Our UK-based client asked us for help with automating manual and time-consuming process of reviewing and approving/rejecting parking ticket appeals.
Brainpool team designed an automated bot which talks to the parking ticket holder to advise them on the local parking regulations in each of the 60 UK regions and makes a decision on whether an appeal should be accepted or rejected based on available decision tree for the specific location.
LLM for automated Press Release Generation
Read moreLLM for automated Press Release Generation
Our client’s team currently spends 3-4 hours a day manually researching articles and writing press release content for their clients in the Life Sciences sector.
Brainpool team designed an LLM-based system that automates the manual content production and generates press releases based on the domain-specific data uploaded for each client account.
The tool is estimated to reduce time spent on manual content production by 50%.
LLM for Customer Services Automation
Read moreLLM for Customer Services Automation
Automation of customer services call centre report processing with Large Language Models.
Our client provides repair services for home appliances in the UK. Brainpool’s team of AI experts developed an LLM-based system that detects what part of a home appliance (e.g. washing machine) is broken, based on the call centre report.
Thanks to this AI-powered analysis an engineer can bring the right part to fix the appliance on the first visit. LLMs improved the accuracy of the client's predictions from 50% to 80%.
LLM-powered Semantic Search for Marketing
Read moreLLM-powered Semantic Search for Marketing
Throughout the Scoping Programme Brainpool team designed a Semantic Search system allowing Client's market research platform to instantly find relevant information within vast amounts of historical data and across various document types
Brainpool data engineering experts are supporting the client's team with implementation of the Al capability within their proprietary marketing platform.
Information Extraction for the big 4 accountancies
Read moreInformation Extraction for the big 4 accountancies
Natural Language Processing can help extract information across a variety of documents such as Word, Excel, PPT, Emails, PDF to maximise work efficiency and document processing speed.
We designed the strategy for a development of an NLP platform for a forensics department at one of the top 4 accounting firms (name remains confidential).
HSBC Project Pulse
Read moreHSBC Project Pulse
Brainpool provided Al expertise for Project Pulse which aimed to build the first cash-flow prediction app for SMEs in the UK. Our PhD-level experts built forecasting models and mobile application components that were implemented into the app.
“We could not be happier with the partnership with Brainpool.”
Ameet Benegal - Director, HSBC Global Commercial Banking
Enhancing Kx Systems with Machine Learning
Read moreEnhancing Kx Systems with Machine Learning
Brainpool team helped Kx Systems data scientists build open-source Machine Learning libraries into one of the most broadly used financial technology in the world called kdb+.
Users were able to call ML models from within their time-series database software, allowing for higher level of automation.
Sentiment Analysis AI Test Drive Platform
Read moreSentiment Analysis AI Test Drive Platform
In 2022 Fujitsu commissioned Branpool to build its AlaaS platform. The first product capability that was incorporated into the platform was a Sentiment Analysis tool that allows brands to measure sentiment towards their offering.
For more information go to:
Demand Forecasting
Read moreDemand Forecasting
Temporal Fusion Transformer (TFT) models are being used to predict item-level demand to a high level of accuracy. Make sure your stock and operational resources are accurately measured and planned for according to the predicted demand
Using TFT we were able to achieve a high level of accuracy of less than 10% MAPE (Mean Absolute Percentage Error) when predicting demand for a US-based restaurant chain.
Machine Vision System to Detect Electric Vehicles
Read moreMachine Vision System to Detect Electric Vehicles
Brainpool team developed a machine vision system that can identify electric vehicles in large datasets of CCTV data from cameras placed o n European motorways.
During the PoC the Brainpool team was able to achieve a higher-than-human accuracy of detecting the electric vehicles by the numberplate.
A network of 500 data scientists working across finance, healthcare, retail, marketing, manufacturing, construction and more.
“We invited Brainpool to participate in the initial definition workshop for Fujitsu's AI ecosystem in Europe. They really brought a level of knowledge and enthusiasm to the session. All stakeholders were thoroughly impressed with their experience in machine learning projects and their business model agility - I would not hesitate to recommend them!”
Leigh Schvartz
Head of Product Innovation, Fujitsu UK