A Indra Energy is developing a platform that, using Generative AI With Amazon Bedrock, it will allow millions of users to migrate to the Free Energy Market.
Operating since 2019, Indra Energia is a company specializing in energy trading and has been supporting clients who wish to migrate to the Free Energy Market.
Thiago Veiga, a partner at Indra Energia, explains that by 2028, this option is expected to be open to all consumers—residential and corporate—which should greatly expand the migration process.
He explains that this process begins with an analysis of the customer's electricity bill, which allows the company to recommend the best product according to that consumption profile.
"Our clients ask us to carry out this process, but we're talking about a market with the potential for millions of migrations," he explains.
Overview
Indra Energia is developing a generative AI-based platform using Amazon Bedrock to automate the analysis of energy bills and facilitate the migration of consumers to the Free Energy Market.
Operating since 2019, the company specializes in energy trading and offers support to clients who wish to switch to this contracting model.
Opportunity – Standardizing energy bills
Therefore, Indra conceived the development of a platform that would allow interested clients to upload their bills and, based on an analysis, obtain automated cost estimates.
“Initially we tried to develop an in-house solution, but analyzing non-standardized information became a challenge,” he explains, noting that, in addition to each dealership having its own layout for the bills, they were received in various formats, such as color and black and white images, text, PDF, among others.
Given this, Indra's data team devised a solution that applies generic data extraction from the different layouts used and employs Generative Artificial Intelligence to make sense of this data, enabling its analysis and automating responses to customers.
"Since Indra's inception, our processes have run on AWS, so we sought information from them to understand how they could help us with this project," he reveals.
“This data input and account analysis process takes us a lot of time, and it’s also very prone to errors. When we’re operating as we expect, with AWS support, this time will be much shorter.”
Thiago Veiga, Partner at Indra Energia
Why AWS – Identifying Data and Semantics
After the initial contacts, AWS indicated that Flexa Cloud for the development of the project, which aims to extract information from different energy bills, standardize it, and provide answers to users.
Indra Energia's data scientist, Ruan Medina, explains that the process of extracting the information was divided into two stages.
“In the first step, we transform everything into text that can be read by the platform, and in the second, the data is identified and given semantic meaning,” he explains, noting that the process allows for the identification of more than 180 different document templates, all with the same data, but presented in different ways.
"AI is fundamental for identifying this data and the semantics of this context," he says.
The idea is to avoid Indra having to collect this data manually for each new client.
“When we saw the possibility of connecting with Flexa, which was already working with Generative AI, and of incorporating a tool into the process with the support of AWS, we found a way to achieve technical development,” he celebrates.
Flexa Cloud CEO Deivid Bitti explains that two layers of artificial intelligence are being used to develop Indra's platform.
The first, based on amazontextIt extracts data from documents in different formats.
“Here, we were able to create a process that extracts and converts the image into text and, in the end, returns a series of data with high accuracy,” he explains, citing examples of the customer's personal data indicated on the bill, energy concessionaire, consumption history and contracted demand, bill details such as fines and charges, tax rates, group marking and tariff classification, consumer unit identifiers, among others.
The second layer includes Amazon Bedrock for the use of Generative Artificial Intelligence, a critical step in the process.
Generative AI enables the identification and facilitates the semantic processing of information extracted from various processes on the platform.
“Bedrock was chosen because it allows us to test a variety of foundation models quickly and at low cost. With Bedrock, we can test different foundation models until we identify the best one,” explains Bitti.
The foundational model Claude-2Anthropic's language was chosen after performance tests showed excellent accuracy for Brazilian Portuguese. The 100-token context window was critical in allowing work with all the data present in the electricity bill, which can have multiple pages.
The architecture also includes Amazon S3 to provide safe and temporary storage for energy bills.
O Amazon Aurora on PostgreSQL It stores all the platform's data in a structured way.
“We are going to transform this system into an API where it will be possible to submit the file and receive the data. It will be a 100% serverless solution, entirely pay-per-use, adapting to demand as it increases or decreases,” he explains.
Results – Preparing the structure for the future
According to Veiga, Indra's use of Generative AI will represent a disruption in the company's processes.
“This data input and account analysis process takes us a lot of time, and it’s also very prone to errors. When we’re operating as we expect, with AWS support, this time will be much shorter,” he says.
The executive emphasizes that Indra is preparing for 2028, when the Free Energy Market will also be accessible to residential customers.
“We are talking about tens of millions of consumers. Just 2% of this market has more than 2 million customers, and to win them over, we would have to analyze around 10 million invoices. We need tremendous robustness to handle this volume,” he compares, noting that today the company takes three to four days to send a quote—which is already faster than many competitors in the market—and that by 2028, this will be done in seconds and via mobile phone.
Indra's CEO, Ingrid Catherine Santos, agrees.
"The market is opening up, and if we don't prepare, we won't be able to meet the demand. This project isn't disruptive in words only."
Ingrid Catherine Santos, CEO of Indra Energia
The executive emphasizes that the solution should bring benefits such as improved customer experience, allowing for high-quality interaction through a self-service platform.
Additionally, the AWS platform accelerates and scales customer acquisition at volume with few manual actions and dramatically reduces the time spent manually reviewing documents for data collection.
Next Steps
Medina goes even further, saying that Indra will have modules to support each step in the journey of its clients and partners.
"What we have here is the module for researching, identifying, and acquiring customers, but we will still expand the frontend for these accesses, managing the benefits received by partners, and implementing various connections to internal and external data for analyzing the data extracted from energy bills."
"There is a whole development demand that will utilize this solution, and we will have a lot to make use of from AWS infrastructure in this process," he predicts.
About Indra Energia
Indra Energia is a Brazilian energy company that, since 2019, has been helping customers reduce costs, make utility bills more predictable, and reduce their CO2 footprint.
