Case Studies

Situation

A 50-year metal-cladding and CNC manufacturer ran production and inventory on manual, disconnected processes. Business data lived in a Priority ERP on SQL Server, but the production data that existed was effectively trapped — there was no way to use it to actually run the floor. There was no engineering team; I was the only developer.

Approach

Over roughly 16 months I built a Priority-ERP-integrated Manufacturing Execution System as the sole engineer: a Next.js/React management app, a real-time Vue 3 shop-floor PWA (live station monitoring, offline support, full Hebrew/RTL for factory tablets), and typed Node/Express services over SQL Server. I led a full JavaScript → TypeScript modernization, integrated CNC sheet-nesting optimization with automated write-back into the ERP, and shipped supporting services — a Python/FastAPI OCR microservice and a React Native barcode-scanner app — plus end-to-end auth/RBAC and Docker + GitHub Actions CI/CD.

Outcome

Adopted across the entire company — roughly 50 people, from management and planners to inventory managers and the shop floor (everyone but accounting). It became the single system of record that made production and inventory measurable and actionable for the first time: live order status, real inventory counts, traceability, and material/CNC visibility the business simply could not see before.

Situation

A real-time order-management platform for a high-volume food-ordering business had to ingest and reconcile data from many disparate sources while staying responsive under heavy order traffic. Manual data-synchronization steps were slow and error-prone, and query/API latency directly affected the live ordering experience.

Approach

Built scalable backend services in Node.js and TypeScript, using MongoDB as the operational store and ElasticSearch for fast lookup, engineered to hold up under load. Designed integrations that kept orders and their supporting data consistent across systems in real time, profiled and optimized the hot database queries and API workflows, and automated data-synchronization tasks that had been manual.

Outcome

Lower response times on the optimized paths, and a recurring source of manual, error-prone work removed through automation.

Situation

A national telecom operator ran core services on C#/.NET with heavy business logic locked inside Oracle PL/SQL. The tightly-coupled stored-procedure logic made services hard to change and slow to onboard new engineers onto, and the delivery process around them was dated.

Approach

As tech lead, migrated C#/.NET + Oracle PL/SQL services to Node.js — deliberately decoupling business logic out of the database rather than lifting it as-is — and standardized the stack on Node.js and Docker for reproducible builds. Introduced Agile and CI/CD to tighten the delivery loop.

Outcome

Better service performance and a codebase the team could onboard into and change more easily, on a more consistent, automated delivery pipeline.