{"id":8763,"date":"2025-04-28T14:13:33","date_gmt":"2025-04-28T17:13:33","guid":{"rendered":"https:\/\/softdesign.com.br\/en\/?post_type=cases&p=8763"},"modified":"2025-09-17T16:19:42","modified_gmt":"2025-09-17T19:19:42","slug":"machine-learning-to-boost-productivity-in-liberum-ratings","status":"publish","type":"cases","link":"https:\/\/www.softdesign.com.br\/en\/cases\/machine-learning-to-boost-productivity-in-liberum-ratings\/","title":{"rendered":"Machine Learning to Boost Productivity in Liberum Ratings"},"content":{"rendered":"\n
\n
\n

Liberum Ratings<\/h4>\n\n\n\n

AI and Machine Learning to boost productivity and enhance the security of FIDC assessments<\/h1>\n\n\n
\n \n
\n \n \n \"Industry\"\n <\/span>\n

\n Industry <\/h4>\n

Finance<\/p>\n <\/div>\n\n

\n \n \n \"What\n <\/span>\n

\n What we delivered <\/h4>\n

Software Development<\/p>\n <\/div>\n\n

\n \n \n \"Since\"\n <\/span>\n

\n Since <\/h4>\n

2023<\/p>\n <\/div>\n<\/div>\n\n\n\n

Liberum Ratings leveraged Data Science and AI to improve its investment fund analysis and significantly increase process efficiency. The transformation encompassed multiple technology initiatives, including data organization, machine learning, and generative AI, resulting in notable gains in productivity and accuracy.<\/strong><\/p>\n\n\n\n

<\/div>\n\n\n\n

The challenge<\/h2>\n\n\n\n

Liberum Ratings<\/a> is the largest credit rating agency in Brazil, holding a 60% market share in the sector. Its investment fund analyses provide qualified insights to support investors in making informed decisions.<\/p>\n\n\n\n

The Backoffice system developed by SoftDesign had already delivered improvements in operational speed and efficiency for Liberum Ratings, cutting data analysis time by 75%. However, as a leading credit risk rating agency, Liberum continued to seek new technologies to manage large volumes of data without compromising the quality of its outputs.<\/strong><\/p>\n\n\n\n

<\/div>\n\n\n\n

Our contribution<\/h2>\n\n\n\n

By implementing machine learning, Liberum Ratings automated report generation and the detection of anomalies within spreadsheets, reducing manual rework and accelerating data validation. Additionally, AI now predicts potential changes in fund ratings, enabling analysts to prioritize critical cases for expert review.<\/p>\n\n\n\n

The most recent advancement was the integration of generative AI to draft analytical reports\u2014previously a manual task for analysts. Now, reports are generated automatically, with analysts responsible only for final reviews, further optimizing work time.<\/p>\n\n\n\n

<\/div>\n\n\n
\n \n \n\n<\/svg>\n <\/i>\n
We used to lose a lot of time analyzing data\u2014we had to open file by file, and it took a long time to generate reports. Depending on the fund\u2019s complexity, it could take two days to complete an analysis; now, I can finish it in one day.” <\/strong>\u2014 Nat\u00e1lia Segato, Senior Risk Analyst at Liberum Ratings<\/blockquote>\n<\/div>\n\n\n
<\/div>\n\n\n\n
<\/div>\n\n\n\n

Error detection in incoming data and fund rating predictions are now powered by machine learning, which has significantly reduced the need for manual validation. The system proactively identifies variations that could impact ratings,<\/strong> allowing analysts to focus on the most complex cases.<\/p>\n\n\n\n

Another major improvement is in data import accuracy. Previously, analysts manually verified data integrity; now, algorithms automatically validate the incoming information.<\/strong> If an error is detected, the import process is halted and flagged for analyst intervention.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Results: greater efficiency and automation<\/h2>\n\n\n\n

Analysts now experience fewer distractions and have more time to focus on critical cases, develop strategies, and conduct deeper analyses. As a result, productivity and analytical capabilities have both improved.<\/p>\n\n\n\n

Through machine learning, Liberum Ratings has achieved:<\/p>\n\n\n\n