The SCAR (Supply Chain Anomaly Radar) system is an operational, data-driven platform developed by the Ministry of Innovation, Science and Technology’s Horizon Line Department in collaboration with the National Emergency Authority. Designed for national resilience, SCAR provides real-time monitoring and predictive early warnings for global disruptions affecting Israel’s most critical strategic supply chains.
Moving beyond theoretical AI prototypes, SCAR implements a Dual-Engine Operational Framework built on a fully automated AWS serverless infrastructure. The first pillar is a Deterministic KPI Alerting System that monitors supply chain health – analyzing price volatility, transport latency, and supply shortages. This engine ingests high-fidelity data from a diverse array of authoritative sources, including UN Comtrade and Israel Customs for trade flows, JODI (Joint Organisations Data Initiative) for global energy balances, and USDA for agricultural production metrics.
The second pillar is an Agentic RAG (Retrieval-Augmented Generation) Workflow powered by Amazon Bedrock. Unlike standard chatbots, SCAR utilizes an Intelligent Agent Router that autonomously orchestrates complex queries. By dynamically selecting the optimal data path whether querying structured SQL tables in Amazon Athena or synthesizing unstructured intelligence from expert reports, news feeds (Trading Economics), and real-time web-scraped data (SERP API) the system provides validated, comprehensive answers to strategic interrogations. Ultimately, the chatbot is architected to exclusively deliver localized strategic advice and actionable mitigation tactics for the Israeli market, translating global macro-volatility into domestic-specific strategic guidance.
By prioritizing a “shipping” mindset over traditional research, SCAR demonstrates how Agentic Orchestration and Multi-Source Data Fusion can transform volatile, high-dimensional global data into a mission-critical radar system for national security.
Ravit Cohen-Segev is a Data Scientist with five years of industry experience, specializing in the application of rigorous analytical research to complex, large-scale data challenges. Although she holds a PhD in Physiology from the Technion, her professional career has been defined by her ability to bridge the gap between deep academic inquiry and the delivery of high-impact business insights. She excels at translating sophisticated scientific methodologies into scalable solutions that drive measurable results.
Keynote session: Hadas Grossmon Ella
Break
Lightning talks session
Roundtable closing
Talk by Hila Paz
Talk by Dr. Moran Mizrahi
Closing remarks
End
Reception
Opening remarks by WiDS TLV ambassadors
Dr. Mor Geva , Tel Aviv University: “MRI for Large Language Models: Mechanistic Interpretability from Neurons to Attention Heads”
Panel: “Pioneering Progress: a strategic look at the GenAI revolution and the new role of data scientists“ Shani Gershtein, Melingo | Mirit Elyada Bar, Intuit | Dr. Asi Messica, Lightricks Moderated by Nitzan Gado, Intuit
Poster pitches
Break
Lightning talks session
Lunch & poster session
Roundtable session & poster session
Roundtable closing
Shunit Agmon, Technion: “Bridging the Gender Gap in Clinical AI: Temporal Adaptation with TeDi-BERT”
Shaked Naor Hoffmann, Apartment List: “Building Generative AI Agents for Production: Turning Ideas into Real-World Applications”
Closing remarks
The end