# Feedzai Welcome to Feedzai Techblog, a compilation of tales on how we fight villainous villains through data science, AI and engineering. - Medium This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Uncovering the Shape of Fraud with Cosmos Explorer: Visual Metaphors Behind Millions of Transactions DevFeed: [Uncovering the Shape of Fraud with Cosmos Explorer: Visual Metaphors Behind Millions of Transactions]() Original publisher: [Read original article]() Author: João Bernardo Narciso Published: 2026-04-07T17:24:51Z Content type: article Language: en Sources: [Feedzai]() Topics: [Data visualization](), [data]() Tags: [data](), [data-visualization](), [dataviz](), [decision-making](), [design](), [feedzai](), [fraud](), [fraud-detection](), [patterns](), [research]() ### AI overview Feedzai's Data Visualization Research team is developing Cosmos Explorer, an interface that uses universe-inspired visual metaphors to help analysts examine patterns, trends, outliers, and possible fraud across hundreds of millions or billions of transactions. The project explores how to preserve meaningful details at very large scale while supporting data analysts and data scientists. ### Source excerpt Uncovering the Shape of Fraud with Cosmos Explorer: Visual Metaphors Behind Millions of Transactions The Data Visualization Research team is developing Cosmos Explorer, an interface that leverages universe-related visual metaphors to convey information about the billions of transactions processed by Feedzai. Pedro Cruz, professor at Northeastern University, partnered with Feedzai to bring this idea to life by contributing with his creativity and expertise to solve this challenging visualization problem. https://medium.com/media/1b2ecfadd91d204640462f89fa6ff67f/href When we look out into the universe, we don't just see emptiness. We see an unimaginable scale: billions of galaxies, each containing billions of stars, each a point of light carrying its own story. No single observer can take it all in at once. Yet with the right instruments, patterns emerge: the structure of the cosmos itself becomes visible. In the digital realm, there is another universe just as vast and intricate. Every day, hundreds of millions of events flow through Feedzai's system which assesses them to protect consumers all over the world. Each one is a unique data point (e.g., a purchase, a login, a transfer). Individually, they don't tell us much. Together they form a living universe of behavior that represents the diversity in people's lives. But fraud lurks in everyday transactions, with criminals trying to hide their activities within the sheer volume of transactions. The question is: how can we represent those patterns meaningfully, the normal behaviors and the fraudulent behaviors, the trends and the outliers, to empower data analysts and data scientists in their decision-making processes? The biggest challenge is scale. No one can look at billions of events one by one. Aggregation helps, but it smooths over the details, which often encode the most interesting signals like the faint outlines of fraud or unusual clusters of activity. But what if we could see it all at once? Not just a summa ## BIY: Preparing a Dataset and Benchmarking AI Models for Scatterplot-Related Tasks DevFeed: [BIY: Preparing a Dataset and Benchmarking AI Models for Scatterplot-Related Tasks]() Original publisher: [Read original article]() Author: João Palmeiro Published: 2026-01-19T14:52:55Z Content type: article Language: en Sources: [Feedzai]() Topics: [Accessibility](), [LLM evaluation / benchmarking](), [multimodal-ai](), [Canvas](), [AI Models](), [dataset](), [SVG]() Tags: [accessibility](), [ai-models](), [artificial-intelligence](), [benchmarking](), [canvas](), [clustering](), [data-visualization](), [dataset](), [large-language-models](), [multimodal](), [multimodal-ai](), [outlier-detection](), [svg]() ### AI overview The article introduces Benchmark It Yourself (BIY), an effort to prepare a dataset and benchmark multimodal AI models for scatterplot-related tasks. It examines generating distribution-aware descriptions from scatterplot images to improve the accessibility of canvas charts, including direct description generation and structured-data extraction for predefined templates. Initial results were mixed. ### Source excerpt Benchmark It Yourself (BIY): Preparing a Dataset and Benchmarking AI Models for Scatterplot-Related Tasks When we need to visualize and interact with millions, or even just thousands, of individual points while analyzing data, we typically resort to rendering them in the browser using a canvas. The other common approach for the web, SVG, doesn't scale when the number of individual elements increases to such quantities. However, while solving one problem, canvas charts introduce a new challenge: accessibility. Although SVG charts are not accessible by default, they can be by design. Each part of an SVG chart has a corresponding element on the web page, allowing for a programmable, accessible experience for screen reader users. We can simply think of SVG as HTML. On the other hand, a canvas chart is just like a PNG image. If a screen reader user tries to learn more about a canvas chart, unless the developer has prepared a detailed description of it, they will just hear the word "image". There's no way to get an idea of what one of these charts represents, let alone extract any insights. For static charts, the solution can be as simple as preparing a description and integrating it into the rendered chart. However, for platforms leveraging dynamic, large datasets, automatically generating these descriptions is not a simple task, especially for charts like scatterplots where data distributions can assume countless forms. At Feedzai, we started exploring ways to generate data distribution-aware descriptions for scatterplots from their respective images using recent multimodal AI models. When the raw data is not available, or the datasets are composed of several thousand or million instances, relying on chart images and these models becomes tempting. This combination has the potential to generate such descriptions and serve them alongside their respective charts, significantly improving the accessibility of canvas charts. That said, we focused on two main directions: using ## Benchmarking LLMs in Real-World Applications: Pitfalls and Surprises DevFeed: [Benchmarking LLMs in Real-World Applications: Pitfalls and Surprises]() Original publisher: [Read original article]() Author: Jean Alves Published: 2025-11-25T15:31:00Z Content type: article Language: en Sources: [Feedzai]() Topics: [LLM evaluation / benchmarking](), [benchmarking](), [Generative AI](), [Security, Privacy and Abuse Prevention]() Tags: [benchmarking](), [computer-vision](), [evaluation](), [financial-fraud](), [fraud-prevention](), [hallucinations](), [large-language-models](), [llms](), [machine-learning](), [trust]() ### AI overview This article explains how Feedzai's ScamAlert uses Generative AI to identify interpretable scam signals from submitted screenshots, and why systematic evaluation and benchmarking are needed to assess consistency, explanations, performance, and limitations such as hallucinations. ### Source excerpt By Jean V. Alves and Ferran Pla Fernández Moving beyond binary classification provides novel insights. In the real world, scams rarely present themselves in black and white. Fraudsters exploit nuance, impersonate legitimate brands, and mask malicious intent with seemingly ordinary behavior. That's why Feedzai has launched ScamAlert (patent pending), a Generative AI-based system innovating on the current paradigm of scam prevention, in response to this growing challenge. Traditional detection systems treat the problem as a binary choice: scam or not a scam, often outputting an estimated "scam likelihood" measure. This value, even if accurate, doesn't tell users why something is risky or what they should watch out for, leaving them with little guidance on how to stay safe. A potential scam SMS The binary approach can often suffer from a lack of context. While a text message may look suspicious in a vacuum (e.g., a payment request via a less safe method) the user may have other reasons to believe in its legitimacy, such as a past history of such requests. Consequently, an incorrect risk estimate based on missing context may lead users to distrust the system's abilities. A traditional binary classification system outputs only a risk estimate ScamAlert, on the other hand, makes judgements on what it knows. Users submit a screenshot of the suspected scam, and ScamAlert identifies observable red flags, patterns or behaviors that are often associated with fraud, such as suspicious links or spelling errors. This approach empowers the user with interpretable insights into the detected risk signals, instead of a vague numeric value. This places the user in the driver's seat, by presenting them with the facts and enhancing their awareness and judgment. To fully understand a systems' ability to perform this task, we pair this labelling approach with a rigorous evaluation and benchmarking protocol. We test the consistency of model outputs for the same input; the model's ability t ## CAUSAL CONCEPT-BASED EXPLANATIONS DevFeed: [CAUSAL CONCEPT-BASED EXPLANATIONS]() Original publisher: [Read original article]() Author: Jacopo Bono Published: 2025-10-03T13:09:07Z Content type: article Language: en Sources: [Feedzai]() Topics: [Machine Learning & Artificial Intelligence](), [Machine learning](), [AI Development]() Tags: [ai](), [blog-post](), [causality](), [concept-learning](), [deep-learning](), [explainability](), [machine-learning](), [research](), [technical]() ### AI overview The article introduces causal concept-based explanations for machine learning models. It argues that existing feature-attribution explanations can be too complex for humans and usually do not support reasoning about hypothetical alternatives. The proposed method aims to provide more human-understandable, reasoning-oriented explanations, with technical details deferred to a publication at the 2024 CLeaR conference. ### Source excerpt Introduction Over the years, we have evolved from using simple, often rule-based algorithms to sophisticated machine learning models. These models are incredibly good at finding patterns in large datasets, but due to their complexity it is frequently challenging for a human to understand why a certain input leads to its respective output. This is especially problematic in areas where high-stakes decisions are being made and where human-AI collaboration is critical. This is why model explainability has gained traction in recent years. The aim of explainability methods is to shed light on what properties of the data contribute to a machine learning model's output. Ideally, such explanations should be similar to how a human would explain its decisions to a peer. In particular, humans make use of high-level concepts that are easy to grasp for other people. For example, a human would explain, "there was a sudden burst of transactions in a short time" as opposed to, "the average delta time feature is much smaller in a recent short time-window compared to a long time-window, and the count of transactions feature increased substantially in the same recent time-window." Moreover, when explaining choices and decisions, humans are free to reflect on these decisions by reasoning about hypothetical alternatives. For example, we could ask, "if this email address would be considered suspicious, would I trust this request?" Current explainability methods fail to incorporate both properties: the explanations are often much more complex than typical concepts humans would use; and they don't allow for reasoning. As a result, these limitations impede the usefulness of current explainability methods in many practical applications. In this blog post, we will describe a new explainability method that addresses these issues. We'll try to refrain from discussing too many technical details and focus on the intuition behind the method, but readers interested in the technical justifications ar ## Feedzai TrustScore: Enabling Network Intelligence to Fight Financial Crime DevFeed: [Feedzai TrustScore: Enabling Network Intelligence to Fight Financial Crime]() Original publisher: [Read original article]() Author: Sofia Guerreiro Published: 2025-07-25T11:59:49Z Content type: article Language: en Sources: [Feedzai]() Topics: [Network](), [Artificial Intelligence](), [AI Models]() Tags: [ai](), [ai-models](), [banking](), [feedzai](), [financial](), [financial-fraud](), [fraud](), [fraud-detection](), [genai](), [machine-learning](), [network](), [network-intelligence](), [report](), [research](), [scams]() ### AI overview This blog post introduces Feedzai TrustScore, part of the Feedzai IQ solution, as an AI-based approach to financial fraud detection. It describes how network intelligence and aggregated knowledge from Feedzai's processed events are intended to help detect threats earlier and adapt to changing fraud patterns. ### Source excerpt By Sofia Guerreiro, Ricardo Ribeiro Pereira, Iker Perez, Jacopo Bono Detecting financial fraud is like finding a moving needle in a shifting haystack. Fraud accounts for a tiny fraction of financial transactions, often less than 0.1%. At the same time, fraudsters are constantly adapting their tactics to evade detection. And this happens within a live and dynamic environment, where financial behaviors and technologies are changing over time. In short, this is an exceptionally difficult problem for financial institutions. With the rise of digital banking and new technologies like GenAI, this problem is becoming even more challenging. In fact, fraud schemes are spreading faster, growing more sophisticated, and becoming increasingly coordinated across geographies. Meanwhile, many financial institutions still rely on detection systems that are either too rigid (e.g., relying on handcrafted rules) or too narrow (e.g., dependent on custom AI models), and neither option can keep up with the pace of change. At Feedzai, we've developed a new approach. In this blog post, we introduce Feedzai TrustScore, part of the Feedzai IQ™ solution. This is a state-of-the-art AI solution that combines the aggregated knowledge from our network of $8.02T yearly processed events to detect threats earlier, adapt continuously, and safeguard financial systems with speed and precision. Traditional Fraud Detection: Catching a Moving Target With a Fixed Net Every year, millions of people fall victim to financial fraud or scams, with a recent report revealing an average loss of over $2,000 per victim. To protect their customers, banks and other financial institutions have made great efforts towards detecting these suspicious transactions, sometimes temporarily blocking them, conducting internal reviews, and contacting the account owner to validate the activity. However, this is an inherently adversarial task, as malicious actors adapt to whatever detection system the banks have in place. These playe ## Here and Now: Reusing Code at Feedzai with JupyterLab Snippets DevFeed: [Here and Now: Reusing Code at Feedzai with JupyterLab Snippets]() Original publisher: [Read original article]() Author: João Palmeiro Published: 2025-03-17T13:11:26Z Content type: article Language: en Sources: [Feedzai]() Topics: [jupyterlab](), [Data Science](), [Code](), [Data analysis]() Tags: [code](), [data-analysis](), [data-science](), [debugging](), [jupyter](), [jupyterlab](), [python](), [repositories](), [snippet]() ### AI overview Feedzai developed JupyterLab Snippets to help data scientists reuse code from the many notebooks created for daily work. The article describes research using notebook repository statistics and user interviews, and reports that duplicating notebooks, maintaining reference collections of snippets, and sharing notebooks or snippets are common practices. ### Source excerpt Data scientists use different Jupyter notebooks every day -- ranging from disposable ones for quick tasks to those shareable with clients. Over time, more and more notebooks accumulate, making it increasingly difficult to reuse them in whole or in part. To mitigate this problem and make the most relevant pieces of code quickly accessible to every data scientist, we developed JupyterLab Snippets at Feedzai -- our take on leveraging code snippets directly on JupyterLab. JupyterLab is a computational notebook platform that enables us to carry on data science work (and beyond) via notebooks. These notebooks, where code, text, and images come together, allow us to complete all kinds of tasks, keeping each input close to each output. At Feedzai, data scientists have access to different JupyterLab environments packed with custom notebooks and Python packages. Here they go from analyzing data to training models, from preparing reports to debugging the system -- a "lab of all trades", we could say. Given the importance of JupyterLab and notebooks in the daily work of our data scientists, we started a research project that culminated in JupyterLab Snippets. First, we collected statistics from internal notebook repositories across teams and conducted user interviews with junior and senior data scientists. We needed to know more about how data scientists actually use JupyterLab and notebooks, and what their ideas are for a better platform. A sneak peek of JupyterLab Snippets. From the very start, every data scientist can import snippets immediately or start their collection by creating one from scratch. After compiling the insights, several aspects became clear: JupyterLab and notebooks are heavily used for all sorts of data science and ad-hoc tasks (typically one per notebook). There is no single type of task (e.g., exploratory data analysis) that is more commonly addressed with notebooks than others. While file structures may look similar at first glance (each environment has at ## "Show Me What's Wrong!": Enhancing Fraud Detection Analysis by Combining Charts and Text DevFeed: ["Show Me What's Wrong!": Enhancing Fraud Detection Analysis by Combining Charts and Text]() Original publisher: [Read original article]() Author: Beatriz Feliciano Published: 2024-11-22T18:35:14Z Content type: article Language: en Sources: [Feedzai]() Topics: [Transactions](), [data](), [synthetic-data](), [Tool]() Tags: [data](), [data-analysis](), [data-visualization](), [financial-fraud](), [fraud](), [fraud-detection](), [fraud-investigation](), [image](), [interface](), [research](), [synthetic-data](), [transactions]() ### AI overview The article describes a fraud-analysis tool that combines charts and text to help analysts review suspicious financial transactions. It explains that tabular review can make it difficult to identify patterns and anomalies within a 1-to-5-minute review window, and presents an interface using synthetic data to help analysts prioritize investigation areas. ### Source excerpt Every year, millions of people fall victim to financial fraud. In 2023, the losses tied to this type of crime were estimated at US$159 billion just in the US, with some people losing all of their retirement savings to scammers. However, the impacts of this issue stretch beyond someone's finances. It can also impact a victim's life in many dimensions. Detecting and quickly acting upon suspicious transactions is essential to tackle this problem. Finding Fraud Through Data Tables To review the data of alerted transactions, analysts look at information in tabular format (similar to what is presented in Figure 1), scrolling through it to assess past activity patterns of the alerted person and comparing those with the alerted event. "How much money was spent on average on past transactions?" or "Is that significantly different from the amount on the current alert?" are some questions they might try to answer during their review. Figure 1: Image of a table that analysts typically use to review the data of alerted transactions. The issue with this approach is that finding groups of patterns and anomalies in tabular data can be overwhelming since it requires an increased cognitive load from analysts to interpret the data effectively. This becomes even more complex since these professionals must review and classify the alerted transaction in a short time -- between 1 and 5 minutes. Revamping the analysis To solve this problem, we present a tool that combines charts and text to guide the analysis of financial transactions. As presented in Figure 2, the tool (populated with synthetic data) is divided into three regions that provide different levels of information detail -- from the most high-level to the most detailed. The goal is that the analyst can scan the charts and prioritize their review towards specific areas of the alert. Figure 2: Proposed interface composed of multiple regions: the Knowledge Area Console (A) to detect suspicious areas of the analysis; the Knowledge Are ## The GANfather: Using Malicious GenAI Agents to Combat Money Laundering DevFeed: [The GANfather: Using Malicious GenAI Agents to Combat Money Laundering]() Original publisher: [Read original article]() Author: Ricardo Ribeiro Pereira Published: 2024-10-04T13:49:42Z Content type: article Language: en Sources: [Feedzai]() Topics: [Generative AI](), [synthetic-data](), [Machine learning](), [AI Development](), [data]() Tags: [ai](), [banking](), [data](), [feedzai](), [financial-sector](), [gans](), [genai](), [generative](), [generative-ai](), [legacy](), [machine-learning](), [money-laundering](), [research](), [synthetic](), [synthetic-data](), [transactions]() ### AI overview Feedzai describes a method that uses Generative AI to create synthetic data simulating realistic money-laundering activity. The generated examples are intended to support machine-learning approaches to detection and help identify vulnerabilities in banks' defenses, addressing the limited availability of labeled data. ### Source excerpt Digital systems have become deeply integrated into many aspects of modern life, particularly within the financial sector. While digital banking simplifies day-to-day operations for clients, it also creates new opportunities for malicious actors to exploit these systems. As a result, money laundering has grown particularly prevalent due to this digital expansion. Banks are required to monitor for money laundering activities and issue alerts when suspicious transactions are detected. Typically, monitoring is performed by rules-based legacy systems. A better approach would be to use Machine Learning models, but these usually require labeled data to train, which are mostly unavailable in this use case. To tackle this problem, we employ advanced Generative AI (GenAI) techniques to generate synthetic data that simulates realistic money laundering activities. These synthetic examples help us identify vulnerabilities and strengthen the defense mechanisms used by banks and other financial institutions In this blog post, we will explore the method developed by Feedzai, which leverages GenAI to tackle the challenges of detecting and preventing money laundering in today's digital landscape. This blog post is the first of a series dedicated to the work done on GenAI by Feedzai Research in the last few years. Problem Statement First, let's briefly introduce the concepts behind money laundering and the difficulties that banks face when trying to prevent it. Money laundering is the process of concealing the origins of illegally obtained funds. Criminals cannot directly spend "dirty" money without risking exposure of their illegal activities. Therefore, they want to disguise the origins of funds before using them. Money laundering typically involves three stages: Placement: the money is introduced into the financial system, often in small amounts spread across various banks. Layering: the money launderer moves the funds through a series of transactions, typically across multiple fin ## Aequitas Flow step-by-step: a Fair ML optimization framework DevFeed: [Aequitas Flow step-by-step: a Fair ML optimization framework]() Original publisher: [Read original article]() Author: Sérgio Jesus Published: 2024-08-12T14:35:43Z Content type: tutorial Language: en Sources: [Feedzai]() Topics: [Machine learning](), [Optimization](), [audit](), [Benchmark](), [pandas](), [datasets](), [interfaces]() Tags: [audit](), [benchmark](), [datasets](), [fairness](), [interfaces](), [machine-learning](), [ml](), [open-source](), [optimization](), [pandas](), [reproducibility](), [research](), [responsible-ai]() ### AI overview This tutorial introduces Aequitas Flow, an open-source framework for standardized experiments with fair machine-learning algorithms. It explains how the framework extends Aequitas fairness auditing with reproducible experimentation, support for custom datasets and methods, and workflows for practitioners selecting models and researchers running benchmarks. ### Source excerpt By Sérgio Jesus, Inês Silva, Pedro Saleiro, Hugo Ferreira, Pedro Bizarro In this blog post we will visit Aequitas Flow, an Open-Source framework designed to run complete and standardized experiments of Fair ML algorithms. We encourage you to try Aequitas Flow with the Google Colab Notebooks, which are available in the project's GitHub repository. This blog post is based on the paper by Sérgio Jesus, Pedro Saleiro, Inês Silva, Beatriz M. Jorge, Rita P. Ribeiro, João Gama, Pedro Bizarro, and Rayid Ghani. Table of Contents: 1. What is Aequitas Flow? - 1.1. For Practitioners selecting a model - 1.2. For Researchers running a benchmark 2. Install Aequitas Flow 3. The components of Aequitas Flow - 3.1. Experiment - 3.2. Optimizer - 3.3. Datasets - 3.4. Methods - 3.5. Audit 4. Conclusion What is Aequitas Flow? Aequitas Flow is the codename for the latest version of Aequitas, a well-established package for fairness auditing in the ML community. This version extends the package to include experimentation with Fair ML algorithms. Aequitas started as a software to diagnose and alert for disparities in ML models' decisions depending on sensitive attributes, such as race, gender, or age. To achieve this, the package runs a Bias Audit, in which it calculates several metrics across all data groups, determined by the sensitive attributes, and compares them to identify any gap in performance. While some other packages already implement Fair ML methods, they also introduce an overhead of technical knowledge to configure, evaluate, and deploy these models. Because of this, we extended Aequitas to enable users to conduct experiments using a wide variety of fair ML methods from the literature in an intuitive user experience. Aequitas Flow was built on the principles of extensibility and reproducibility: it allows users to incorporate their own datasets and methods in the framework using familiar interfaces of sci-kit-learn and pandas. It ensures their work can be replicated by performin ## Building Trust in a Digital World: The Role of Machine Learning in Behavioral Biometrics DevFeed: [Building Trust in a Digital World: The Role of Machine Learning in Behavioral Biometrics]() Original publisher: [Read original article]() Author: Javier Liébana Published: 2024-06-21T14:01:53Z Content type: article Language: en Sources: [Feedzai]() Topics: [Digital Trust](), [Machine learning](), [Machine Learning, Security Attacks]() Tags: [core](), [digital-trust](), [feedzai](), [fraud-detection](), [fraud-prevention](), [machine-learning](), [online-fraud-prevention](), [research]() ### AI overview This article explains how Feedzai's Digital Trust solution uses behavioral, device, network, operating system, and browser data to support digital identity verification and transaction authorization. It describes a machine learning model that continuously evaluates collected data and uses insights from previous fraud to improve fraud detection and prevention. ### Source excerpt In the world of financial services, the bank or financial institution's relationship with the customer relies on digital trust, which is anchored in two fundamental principles. First, it must ensure the person engaging through digital banking channels is genuinely the individual they claim to be. Second, it must confirm that this person is authorized to complete the intended financial transaction. Addressing these crucial requirements is the core mission of Feedzai's Digital Trust solution. The solution collects and analyzes comprehensive user behavioral data, scrutinizes device information for potential threats, such as malware attacks, and evaluates contextual factors like network, operating system, or browser information to gain a complete understanding of the user's environment. However, the high volume and heterogeneous nature of the collected data, among other challenges, makes detecting potential fraudulent sessions with high accuracy a formidable endeavor. In this blog post we explore how a new machine learning (ML) model that performs a continuous evaluation of collected data and leverages insights from previous frauds to vastly improve Digital Trust's fraud prevention capabilities. We will start with an introduction to technical details behind our Digital Trust solution, going into the challenges of fraud detection and prevention. We continue by explaining how we can apply ML to boost fraud detection and how we deployed the new Fraud model to dozens of Feedzai customers. Table of Contents 1. Digital Trust data collection - 1.1 The user journey - 1.2 Behavioral biometrics data - 1.3 User's behavior - 1.4 Device and network data 2. Challenges to detect fraud in Digital Trust 3. Machine Learning for Digital Trust - 3.1 The holistic approach - 3.2 New Fraud Model 4. Deploying the model 5. In summary Digital Trust data collection To better identify the challenges that are typically faced when designing a fraud prevention system based on Digital Trust, first we