Arteris, a leading highway management company responsible for 2,000 miles of road in Brazil and handling approximately 672 million vehicles annually. Arteris leadership recognized the potential in AI and decided to make their vision for the future of intelligent highway management a reality.

To bring this vision to life, Arteris initially approached Microsoft. Recognizing the specialized computer vision and AI expertise needed for such a complex undertaking, Microsoft encouraged a strategic partnership with Programmers Inc.. Microsoft was confident in our ability to tackle this vital challenge and architect a solution on the Databricks platform.

Their vision for the future:

Arteris’s goal was to create a new generation of highway management systems, and have AI at the core of that vision. Some of the features of this new system would be observability of what is going on across their roads, and to make real-time decisions to make the roads better.

The Situation:

Arteris’s existing method of data collection relied on outdated and inefficient sensors which needed to be embedded in the pavement. While providing some information, they were expensive to maintain, offered slow time-to-analytics, and, crucially, the data lacked the necessary accuracy for transformative decision-making.

Our Solution: AI-Powered Computer Vision for Real-Time Insights:

Programmers Inc. embraced the challenge of revolutionizing Arteris’s data gathering capabilities. Our team developed and trained a sophisticated computer vision model leveraging Arteris’s existing camera network. This new system surpassed the legacy sensors by:

• Increase traffic data accuracy while providing data in real-time.
• Identifying a wide range of hazardous situations previously undetectable by sensors, including fires, broken-down vehicles, pedestrians on the roadway, and animals.
• 80% cost savings when at scale
• Less of a burden on operations

A significant problem-solving hurdle was the limited availability of training data for specific scenarios, such as identifying various animal types near roadways. Programmers Inc. demonstrated its innovative approach by employing GenAI to augment the training dataset – for example, by digitally adding images of horses to roadside scenes – thereby significantly improving the model’s accuracy and versatility. This creative solution was key to overcoming a critical data gap.

The “Fast To Value” approach utilized by the Programmers Inc., was vital to the project’s success. Each stage, or sprint, of the project was planned in 90-day cycles, bookended by a rigorous approval process. This iterative approach allowed for continuous review, adaptation to evolving project needs, and incorporation of stakeholder feedback, ensuring the solution remained aligned with Arteris’s goals. Steady release schedules with well-defined features facilitated smooth adoption by end-users, enabling them to realize immediate benefits from each system update, further solidifying their confidence in Programmers Inc.’s delivery.

The newly developed AI system has delivered transformative results:

• Significant Cost Savings: Potential for up to 80% cost savings on vehicle counting and monitoring equipment, as highlighted by Arteris’s CTO.
• Accelerated Analytics: Data processing and analytics are now 3 times faster than the old system.
• Enhanced Data Quality: The system provides more comprehensive data with higher accuracy, leading to better-informed decision-making for road safety interventions.
• Removed data silos

“I would like to thank and congratulate the entire Programmers team for their engagement, collaborative attitude, and partnership, which created a light and fun work environment. I also highlight all the technical competence and the pursuit of new learnings in order to deliver more assertive solutions. Thank you very much, and see you next time!”

The Future: Expanding Impact and Continuous Innovation

While the new AI system is currently active in select areas, a nationwide deployment across Arteris’s network is planned over the next two years, pending further approvals. The Programmers Inc. team continues its dedicated work, committed to solving future challenges and evolving the system. Upcoming enhancements aim at enabling the system to interact autonomously with dispatch services, helping resolve hazards quicker

_____________________________________________________________________________________________

Reach out to Programmers Beyond IT today to explore how our AI and data-driven expertise can drive tangible business results for you.

A new application developed by Programmers is facilitating the technical consultancy of a global bearing manufacturer, allowing it to quickly identify which spare parts are compatible with its competitors’ products. The result is optimized customer service and increased sales. 

First, imagine that a part of industrial, automotive or even household machinery has broken and you need to find a replacement. With a huge variety of models available on the market, each with its own technical specifications, the replacement process can be challenging. In this way, without a clear way to compare the alternatives, you end up restricted to buying the same brand already in use. 

This was the scenario faced by the team. In short, engineers constantly need to inform their customers which parts they produce can replace competitors’ parts. 

Challenge 

Given the wide diversity of bearings, it was a challenge to make this comparison and indicate which part was ideal to replace the competition’s. Therefore, the company’s main challenge was the speed and accuracy required in the process of identifying compatibilities. 

Engineers needed a tool capable of performing technical comparisons in an agile manner. The lack of an integrated solution generated delays, impacting customer response time and making it difficult to convert sales. 

Solution 

Faced with this scenario, Programmers developed a customized web application and a specialized database that maps the correlation and compatibility between the company’s own bearings and those of its competitors. The tool is designed to enable the engineering team to quickly identify compatible parts, saving time and improving the accuracy of recommendations made to customers. 

In addition, the solution offers flexibility: new parameters and products can be added to the database on an ongoing basis, ensuring that the application is always up to date with the latest releases. After the initial delivery of the solution, the application continues to evolve, with the implementation of new features in a new phase of the project. 

Application Results 

The application generated relevant impacts for the bearing manufacturer, including: 

Cost reduction: The time to search for and analyze compatibilities has been drastically reduced, allowing the engineering team to streamline the consulting process; 

Improved customer experience: With a more agile and effective solution, the identification of spare parts has become more efficient, increasing customer satisfaction and strengthening the relationship with the brand; 

Increased sales: Customers who previously used competitors’ bearings are now able to easily migrate to the company’s products, expanding market share and customer loyalty. 

Custom Application Development 

At Programmers, our commitment is to deliver tailored solutions that go beyond the simple implementation of technology. We analyze our clients’ processes, needs, and goals to create tools that generate real value and directly impact business results. 

Contact us and find out how we can leverage the results of your business! 

_____________________________________________________________________________________________

Reach out to Programmers Beyond IT today to explore how our AI and data-driven expertise can drive tangible business results for you.

A large global company needed to optimize internal HR and Infrastructure service, reducing the high volume of Level 1 (L1) calls — those that were simpler and more recurring, but still consumed a lot of support team time. Overloading teams and rising operating costs required an efficient solution. The answer? A corporate chat with Generative AI, developed to streamline service, improve the employee experience, and optimize processes. Check out how our AI solution transformed internal support and brought expressive results!

Challenge

The main challenge was to deal with the high demand for L1 tickets and the inefficiency in accessing the ServiceNow knowledge base. But without automated triage, analysts faced rework and the employee experience was compromised.

In addition, the absence of an intelligent mechanism to classify tickets increased the average resolution time and impacted the productivity of the support team. The company needed a solution that would automate service and improve operational efficiency.

Generative AI Solution

Programmers implemented a corporate chat with Generative AI, fully integrated with ServiceNow, to optimize internal support. The solution included:

Chatbot with Generative AI: Quick and intuitive access to the knowledge base, reducing the need for human interaction.
Automated triage: Intelligent sorting of tickets into “Question” or “Request,” ensuring that only relevant tickets were routed to human service.
RAGAS Framework: implementation of metrics to continuously evaluate and improve AI performance.
Omnichannel Integration: support via WhatsApp, Web and Live Agent, ensuring a fluid experience for users.
Process automation: direct integration with the HR and Infrastructure databases, allowing actions such as password change and user unlocking.

In this way, AI took over repetitive tasks, allowing the support team to focus on strategic and more value-added demands.

Findings

Even in its initial phase, the solution has already generated significant impacts:

10% reduction in the volume of monthly calls, reducing operating costs.

Easier access to the knowledge base, improving efficiency in resolving queries.

Decrease in average resolution time, allowing employees to solve problems autonomously.

In the pilot phase, the numbers were promising:

The chat recorded 1331 unique users and 4592 initial interactions, growing to 2876 users and 9531 interactions on Go Live.

44,538 questions were processed, with 20,266 actions performed and 24,272 responses generated.

WhatsApp had 1776 active users, while the Web version had 1100 users.

Significant reduction in human interactions in service channels: Significant reduction in human interactions in service channels: 84% in Infra and 98% in HR.

Improvement in the IVR (Interactive Voice Response Unit) service rate – an automated system that directs calls and responds to queries without human intervention, rising from 87.5% to 95%, and a drop in the rate of abandoned calls from 500 to only 69.

Generative AI to Drive Care

In this way, the implementation of corporate chat with Generative AI was a milestone in the company’s digital transformation, bringing efficiency, cost reduction, and improvement in the employee experience.

With the expansion of the solution, the gains obtained will be even more expressive, consolidating an agile and intelligent internal support.

So, do you want to know how we can transform your company’s service with AI? Talk to our experts!

_____________________________________________________________________________________________

Reach out to Programmers Beyond IT today to explore how our AI and data-driven expertise can drive tangible business results for you.

Digitalization is not a new concept; Companies have been under pressure to modernize their operations for more than a decade. Organizations have already gone through several waves of technological transformations, such as the arrival of the internet, cloud storage, and the use of mobile devices. Now, AI emerges as the next step in that journey — but also as a challenge that redefines the rules of the game.

Our Operations Director BR, Rafael Dourado explains how AI has changed the scenario. “What arrived as a trend, has now become a matter of survival. The paradox is that, although it was developed to drive digital transformation, this technology has also cornered companies in a single direction: innovation,” says Dourado.

Urgency of AI

In the past, many organizations have embraced modernization cautiously, doing only the essentials to avoid obsolescence. Developing a website or migrating some files to the cloud were enough steps to meet the need to “be up to date”. However, the arrival of AI has put companies at a crossroads: those who do not adopt this technology risk disappearing from the market.

According to a Gartner study, 75% of global organizations plan or are already implementing AI solutions by 2025. In addition, by 2026, 85% of interactions between customers and businesses will be managed by AI. These numbers highlight the urgency of integrating technology, pressuring companies to modernize their processes in a robust and efficient way.

Legacy Systems and Modernization with AI

One of the biggest challenges of modernization is compatibility between AI and legacy systems, many of which still operate with low-quality or improperly stored data. This mismatch directly impacts the efficiency and outcomes of AI-based initiatives.

Legacy systems often lack adequate structures to handle, organize, and integrate data in a way that is useful for artificial intelligence. Data stored in a fragmented, duplicated, or non-standardized way makes it difficult to generate strategic insights. Data analytics, therefore, is the building block that ensures that the implementation of AI delivers tangible value to the business.

Accelerating modernization with AI

Despite the challenges, there is a solution: AI itself can speed up the modernization process. Our Customer Success Manager, Railson Fernandes explains that this technology is transforming the traditional approach to modernization.

Typically, a team would be dedicated for an extended period of time to analyzing the complexities of legacy systems, planning the migration, and ensuring that nothing gets lost in the process. But with AI, it is possible to accelerate all this. It helps you analyze the system, understand the layers of complexity, and even suggest migration solutions. And it does it intelligently, documenting every step.

AI not only simplifies technical processes, but also transforms functionalities, creating richer and more personalized experiences for users. “I think of modernization from two perspectives. Use AI to accelerate development, but also to bring innovation to my application. For example, I’ll understand my user’s profile and create more personalized experiences. It is not only shortening the development cycle, but also adding value to the end user”, bets Railson.

End-to-end modernization

Artificial intelligence is revolutionizing end-to-end modernization. It accelerates processes, improves operational efficiency, introduces new functionalities, and transforms the customer experience. For businesses that want to stay competitive, adopting AI is not just a strategic choice, but an urgent need. Discover our App Modernization offering and transform your legacy systems into modern and efficient platforms. Contact us and find out how we can drive your company’s digital transformation

When we think we are getting good at something, new technology emerges, which makes us go back to studying. But this time it’s a little easier, since it’s a subject that I like a lot: Database! If you also like this theme and want to learn the basics of vector database… This article is for you!

What is a vector database?

A vector database is a category that indexes and stores embedding vectors, providing an efficient search. These databases have the ability to save, modify, delete, and recover data, offering an innovative approach to information management.

What is an Embedding?

Embeddings, in turn, are vectors of texts generated by AI models. These have several characteristics, making their representation a challenge due to their complexity. In contexts such as artificial intelligence and machine learning, these characteristics represent different dimensions of data, which are essential for discerning underlying patterns, relationships, and structures. In summary, it is a way of storing data and its meanings, giving it a unique semantic dimension.

Why is this important?

The representation of vector data, enriched with semantic information, is crucial for artificial intelligence to gain understanding and maintain a long-term memory. This proves to be fundamental in performing complex tasks, allowing AI to extract valuable insights from the vast set of data with which it interacts.

But where am I going to use it? Imagine the scenario in which you are developing artificial intelligence for your company, requiring it to keep crucial information in its memory, such as policies, product data, prices, customers, among others. Therefore, to give this AI an efficient memory, the choice for a vector database becomes essential.

Image generated by DALL-E, depicting a woman giving memory to a robot.

Where to start?

To take the first steps in this universe, choosing the right database is crucial. Among the available options, some stand out:

Pinecone: Developed by a startup, it is easy to learn, allowing free trials.

Qdrant: An open-source option, requiring the configuration of a docker for local execution, but it offers a full range of features.

Azure Search: An excellent choice for Azure users looking for an enterprise option.

Redis: Recognized for its undeniable capacity, Redis, in addition to its applications in other data structures, also offers a vector database.

In summary, vector databases represent a significant evolution in the field of information management. Their ability to represent data semantically and efficiently makes them an indispensable tool for developing robust artificial intelligence systems. By choosing the right vector database and understanding the importance of this approach, we are prepared to enter the future of data, empowering our technological creations to reach new heights of understanding and performance.

_____________________________________________________________________________________________

Rafael Dourado is BR Operations Manager at Programmers. Its main role is to ensure end-to-end customer satisfaction to embrace this technological revolution. With 15 years of experience working in the software and analytics universe, he is passionate about AI because he believes in its transformative power. In his leisure time, Rafael likes to practice chess with his daughter.

The intersection between academia and the job market in the area of data science and machine learning is wonderful — and complex. While in personal projects and competitions we are invited to focus on developing innovative algorithms and obtaining impressive and accurate scores, in the business environment, the challenges go far beyond simply building models.

Some recent readings have awakened this reflection on the main differences between the construction of algorithms in academic contexts/personal portfolios and their construction within large corporations. Software engineering can play a crucial role in the practical implementation of ML systems.

To better understand these differences, I turned to the excellent book “Designing Machine Learning Systems”, by Chip Huyen. It offers valuable insights into the complexities involved in implementing machine learning systems in production environments.

Scalability versus maintenance

One of the key differences is the emphasis on scalability and maintenance of systems that happens in the job market. On the other hand, there is the constant search for metrics optimization, which usually dominates the academic/competition/portfolio scenario.

While the challenges posed in competitions like Kaggle often involve honing a single model to maximize accuracy, or another performance metric, in the data science market professionals are faced with the task of integrating these models into larger systems, dealing with concerns such as scalability, computational efficiency, solution explainability, and long-term maintainability. This means that, yes, a few percentage points of accuracy are often given up, or even the choice of the model itself, for a more fluid integration with the system as a whole.

Individual versus team

The book also underscores the importance of interdisciplinary collaboration and effective communication in the workplace. Data scientists often need to collaborate with software engineers, business analysts, and other professionals to develop solutions that meet the needs and constraints of the business.

This collaboration contrasts sharply with the more ‘individualistic’ nature of the academic or competitive environment. This is because participants often work independently to develop their solutions.

Data Maintenance

However, one of the most significant differences between the academic approach and professional practice is in the stage of preparing and maintaining the data. In personal projects or challenges, often the focus is mainly on the modeling itself. Much of the time and resources are dedicated to experimenting with algorithms, feature engineering techniques, and hyperparameter optimization. Therefore, this approach tends to underestimate the importance of data quality, and all the time spent on its preparation, labeling, validation with business, and the like.

But it is worth remembering that the need to ensure data quality and consistency is not just an initial obstacle. It persists throughout the entire lifecycle of the ML system. A particularly relevant aspect is the concept of “model drift”. It refers to the deterioration of model performance over time due to potential variations in the input data. These changes can be caused by a variety of factors, such as changes in user behavior patterns or changes in operating conditions. Dealing with model drift requires continuous vigilance and effective monitoring strategies, as well as proactive maintenance of models in production.

Software Engineering

In summary, it was thinking about all these differences, and after working the last entire year on a project focused on maintenance and monitoring of models, that I chose to do a specialization in Software Engineering. The ability to understand and apply HE principles in machine learning projects allows this look beyond simple model building and addresses the broader challenges associated with implementing AI solutions in real-world environments.

One of the main advantages of delving deeper into the Ops universe is the ability to develop more robust and scalable ML systems. Deepening knowledge in software development practices broadens the vision. It also enables the adoption of a more structured and modular approach in the design and implementation of data pipelines and models. This makes systems easier to maintain and scale as business demands can evolve.

Data Science Skills

Today we have the role of the Machine Learning Engineer. Deeply immersed in data and software engineering, he brings this look to the implementation processes and MLOps in a much more active way than the Data Scientist.

However, I reinforce that these are skills that all Data Science professionals would benefit from having. They provide a more comprehensive understanding of testing best practices and continuous integration, and continuous delivery (CI/CD). And this is essential to ensure the quality and reliability of the model. They also broaden the look in terms of trade-off costs x scalability, ease in maintaining the life cycle x complexity, and even in the users’ understanding of how the solution works. Depending on the business, this may be the most important metric.

This is knowledge that makes you more capable of developing models in a useful and absolutely viable way for the company to implement them. And at the end of the day, that’s about it.

_____________________________________________________________________________________________

Leticia Gerola is a Data Science specialist at Programmers. His main responsibility is to lead and implement Machine Learning projects in a wide range of clients from various market sectors through cloud technology. With a career transition from journalism to the data area, he has a passion for new projects involving Artificial Intelligence and MLOps. In her leisure time, Leticia surfs on the beaches of the coast of São Paulo and participates in capoeira circles in the capital.