Ravit Cohen-Segev
Horizon Scanning Lab - Data Scientist
Haifa, Israel
Actions
I am a Data Scientist with 5 years of industry experience. Although I hold a PhD in Physiology from the Technion, my professional career has focused on applying the rigorous analytical methods of a research scientist to complex, large-scale data challenges. I pride myself on my ability to bridge the gap between academic inquiry and delivering high-impact business insights.
Area of Expertise
Topics
Building SCAR: Low-Cost, High-Throughput Supply Chain Ingestion and Agentic RAG
When global supply chains fracture, decision-makers cannot wait for manual analysis. Developed alongside Israel’s National Emergency Authority, SCAR is an operational radar providing real-time monitoring and early warnings across critical national supply chains.
This lightning talk breaks down SCAR’s Dual-Engine Operational Framework, built entirely on AWS serverless architecture. We will explore how we pair a deterministic KPI pipeline—utilizing Amazon Athena and S3 for low-cost, high-throughput global trade data ingestion with an Agentic RAG chatbot powered by Amazon Bedrock. Specifically, we focus on the design of our Intelligent Agent Router, which autonomously splits queries across structured SQL tables and unstructured intelligence feeds.
Attendees will walk away with a production-ready pattern for routing hybrid data streams to deliver localized, actionable strategic guidance under crisis conditions.
SCAR: A Multi-Engine Supply Chain Anomaly Radar Driven by Agentic RAG and Real-Time Automation
Abstract: 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.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top