Ori AbramovskyinTowards Data ScienceClassifying Source code using LLMs — What and HowSharing our experience at making LLM-based Source Code classifier14 min read·Dec 28, 2023----
Ori AbramovskyinTowards AI3 years of Google Colab usage — The good, the bad and the uglySummarising 3 years of extensive Colab use8 min read·May 12, 2023----
Ori AbramovskyinTowards AITaking into account temporal aspects of Machine Learning appsTemporal features require special handling, for some cases it can even mean to overfit on purpose8 min read·Jan 19, 2023----
Ori AbramovskyCheck Point CloudGuard Spectral exposes new obfuscation techniques for malicious packages on PyPIPiPY malicious package uses Steganography to hide its intent and GitHub open-source projects to infect users7 min read·Nov 10, 2022----
Ori AbramovskyinTowards AISteps Toward MLOps Research — Software Engineering Your AIKey concepts to make sure your AI developments are Production ready8 min read·Jul 21, 2022----
Ori AbramovskyinTowards Data ScienceThe inherent Train-Inference mismatch of SourceCodeAIWhat to do when we train on X, inference on Y, and measure on Z7 min read·May 25, 2022----
Ori AbramovskyinTowards Data ScienceAnalyzing customers state using spatial Anomaly DetectionWhat to do what only unlabeled data is available8 min read·Apr 27, 2022----
Ori AbramovskyinTowards Data ScienceSourceCodeAI — Deep Learning for Source Code — Why and HowYet another NLP? No No and No9 min read·Mar 28, 2022----
Ori AbramovskyinCodeXDo data practitioners are the new (security) weakest link?Secrets in code5 min read·Mar 2, 2022----
Ori AbramovskyinTowards Data ScienceQuestion pivoting —Handling unachievable AI tasks3 possible pivots when a classification model training is not possible5 min read·Feb 7, 2022--1--1