January 2026

Active learning (AL) strategies aim to train high-performance models with minimal labeling efforts, only selecting the most informative instances for annotation. Current approaches to evaluating data informativeness predominantly focus on the data’s distribution or intrinsic information content and do not directly correlate with downstream task performance, such as mean average precision (mAP) in object detection. Thus, we propose Performance-guided (i.e. mAP-guided) Reinforced Active Learning for Object Detection (MGRAL), a novel approach that leverages the concept of expected model […]

Ver mais

From “rug pulls” to corporate warfare, the rules of software ownership are being rewritten. But a new model—Post-SaaS—might finally offer a way out. For decades, the deal was simple: developers gave their time, and companies gave their code. It was an unwritten social contract built on trust. Then, in a few short months, the contract was shredded. First came the tremors. In August of 2023, HashiCorp announced that Terraform—the industry standard for infrastructure-as-code—was switching to a “Business Source […]

Ver mais

Transformer-based models, despite their promise for long-term time series forecasting (LTSF), suffer from an inherent low-pass filtering effect that limits their effectiveness. This issue arises due to undifferentiated propagation of frequency components across layers, causing a progressive attenuation of high-frequency information crucial for capturing fine-grained temporal variations. To address this limitation, we propose Dualformer, a principled dual-domain framework that rethinks frequency modeling from a layer-wise perspective. Dualformer introduces three key components: (1) a dual-branch architecture that concurrently models […]

Ver mais

Primary bottleneck for Enterprise AI is not the availability of tools or the identification of a tech stack, it is getting the data landscape in order. Success in 2026 is predicated on having total clarity of the underlying data infrastructure and establishing a foundation that is petabyte-scale, secure, and high-performing. Without a reliable data layer, AI initiatives remain experimental rather than transformational. Foundation (Scalable and Maintainable Data Acquisition) A useful litmus test for the engineering foundation is time […]

Ver mais

Modern businesses run on digital trust. Customers expect their data to be safe, and partners expect operations to be reliable. To meet those expectations, security must be a daily practice, not a yearly project. This article walks through practical measures that reduce risk fast. Each section focuses on actions you can apply in most environments. The goal is simple – shrink the attack surface and strengthen your defences without adding needless complexity. Assessing Your Risk Landscape Start with […]

Ver mais

This is an opinion piece based on my research and ideas. I recently read the paper by Alex Kontorovich, The Shape of Math To Come, which inspired me to contemplate the future of mathematics and mathematicians. However, I will not go into as much detail, but rather state a high-level overview of ideas. Autoproving vs Autotranslation Before we start, we need to differentiate between autoproving, giving a formalized proof to a formalized statement, and autotranslation, the act of […]

Ver mais

In quantitative trading, it’s important to be careful when testing your strategy on all available data, as this can sometimes cause the rules to become too tuned and not perform well in real trading situations. To get a better idea of how your strategy really works, try dividing your historical data into a training (or optimisation) period and a test period. This approach allows you to evaluate your strategy on unseen data, much like real market conditions, which […]

Ver mais

For years, our data visualization toolbox in Python has been dominated by two giants: Matplotlib and Seaborn. They are the undisputed champions of static, publication-quality graphics. They allow us to create beautiful, precise “photographs” of our data. But in the modern world of web-based dashboards and dynamic reports, a photograph is often not enough. Static charts are a one-way conversation. What if your charts could talk back? What if your users could hover over a data point to […]

Ver mais

Realistic network traffic simulation is critical for evaluating intrusion detection systems, stress-testing network protocols, and constructing high-fidelity environments for cybersecurity training. While attack traffic can often be layered into training environments using red-teaming or replay methods, generating authentic benign background traffic remains a core challenge — particularly in simulating the complex temporal and communication dynamics of real-world networks. This paper introduces TempoNet, a novel generative model that combines multi-task learning with multi-mark temporal point processes to jointly model […]

Ver mais

I love Excalidraw for sketching system architectures. But sketches are static. When I want to show how a packet moves through a load balancer, or how a database shard splits, I have to wave my hands frantically or create 10 different slides. I wanted the ability to “Sketch Logic, Export Motion”. The Goal I didn’t want a timeline editor (like After Effects). That’s too much work for a simple diagram. n I wanted “Keyless Animation”: Draw Frame 1 (The start state). Clone […]

Ver mais