Shubham Sharma

Built an AI agent–driven anomaly detection and resolution system, automating carrier/driver interactions and reducing manual investigation effort by ~30–40%. Developed backend pipelines to process real-time shipment events and apply automated updates, improving anomaly resolution time by ~25% across multiple use cases.

Software Development Engineer at Project44 (2026-03 – Present)

• Built an AI agent–driven anomaly detection and resolution system, automating carrier/driver interactions and reducing manual investigation effort by ~30–40%.
• Developed backend pipelines to process real-time shipment events and apply automated updates, improving anomaly resolution time by ~25% across multiple use cases.

Software Development Engineer at Esko (ArtworkFlow) (2025-06 – 2026-02)

• Designed and built a scalable, context-aware Notification Service in Spring Boot powering ArtworkFlow's event-driven communication pipeline saving over 9600 USD anually.
• Implemented a dynamic multi-tenant architecture using @ConditionalOnProperty, enabling automatic switching between PostGres and Esko-specific repositories mainly DynamoDb based graph like repositories, templates, and event handlers.
• Refactored legacy monolithic DB access into clean Repository Wrapper Services, reducing cyclic dependencies and improving maintainability across 15+ services.
• Developed a pluggable Notification Factory Framework to route notifications across email, in-app, and SuprSend channels without modifying business logic.
• Integrated Feign-based S2 client with authentication, proxy configuration, and fault-tolerant retries, improving external dependency reliability by 40%.
• Enhanced performance by optimizing JPA queries, adding caching layers, and reducing redundant fetches—cutting template rendering latency by 25%.

Software Development Engineer at Fiorano Software (2024-06 – 2025-06)

• Built a modular backend framework to track latency metrics for different messaging schemas (Pub/Sub, PTP), improving system observability and performance diagnostics.
• Led the end-to-end design and development of an automated ticket intelligence system integrated with Zendesk, improving customer support efficiency.
• Developed a backend pipeline using Python, Asyncio, Groq, Deepseek AI to fetch ticket threads via the Zendesk API.
• Implemented LLM-powered conversation summarization, semantic categorization, and iterative context refinement.
• Built secure connectors to publish internal notes and automate ticket triage, significantly reducing manual workload and improving resolution prioritization.

M.Sc(Hons) Economics and Bachelors Of Engineering Electronics And Electrical Engineering in Economics and Electronics And Electrical Engineering – BITS PILANI KK BIRLA GOA CAMPUS (2019 – 2024)

Central Board Of Secondary Education – Sacred Heart Senior Secondary School Kota (2017 – 2019)