Public proof across product engineering

Case studies

• Agentic systems · local-first product

  Norn agentic finance analysis

  Personal finance context is scattered across documents, statements, and local knowledge, while trust depends on privacy and read-only safety.

  Built: A read-only agentic chat experience for Israeli household finance with document import, anonymized extraction, charts, and finance-aware reasoning.

  Impact: Turns messy financial context into an understandable conversational product without making privacy feel like an afterthought.

• Azure Monitor · observability agent · Microsoft

  Observability Agent in Azure Monitor

  Azure Monitor users need to understand service health across logs, alerts, metrics, and resource context without turning every question into a manual telemetry hunt.

  Built: AI-assisted agent experiences in Azure Monitor that connect observability signals, reasoning flows, automation, and UI/API pieces into guidance that points to a confident next action.

  Impact: Helps teams move from raw signals to plain-language explanations and actions with less repetitive manual troubleshooting.

• Telemetry · automation · product insight

  Operational intelligence and reasoning

  Operational data is only useful when it points people toward the right question and the next useful action.

  Built: Signal processing, alerting, dashboards, feedback loops, and automation around production data.

  Impact: Makes troubleshooting flows easier to prioritize and less dependent on repetitive manual checks.

• Next.js · Three.js · product storytelling

  Interactive 3D career portfolio

  A traditional resume does not show product taste, frontend architecture, graphics performance, or accessibility choices.

  Built: A Next.js and Three.js career timeline with realistic textures, progressive loading, keyboard navigation, accessible overlays, and selected-work flows.

  Impact: Makes the career story memorable while proving product engineering, frontend quality, and performance discipline in the artifact itself.

• Enterprise web systems · Elbit Systems

  Full-stack production delivery

  Enterprise products need someone who can move across the UI, API, data, and production-support seams.

  Built: Responsive UIs, backend integrations, performance fixes, and support/debugging improvements.

  Impact: Built the full-stack habits that later translated well into cloud-scale product engineering.

• Code review · mentoring · alignment

  Engineering leadership and mentorship

  As teams grow, quality depends as much on shared habits as on individual output.

  Built: Practical documentation, review feedback, and technical alignment around trade-offs.

  Impact: Helped engineers move with more confidence while preserving readability and operability.