Rohan Shah

Conducted rigorous testing and evaluation of advanced AI models to optimize reasoning capabilities and align them for more accurate responses. Compared two high-class models via public repository cloning and critical prompting to evaluate performance.

Led design and architecture of an asynchronous, multi-stage data orchestration engine containerized via Docker and deployed on AWS EC2 instances, executing 150K+ daily automated data pipelines with 99.95% system availability. Developed an event-driven AI reasoning engine chaining Claude 3.5 Sonnet and GPT-4o REST APIs via custom Node.js and Python FastAPI mid-layers.

Architected an enterprise-grade logistics automation platform, transitioning legacy, fragmented supply chain operations into a decoupled microservices architecture orchestrated via Kubernetes. Designed a high-throughput data synchronization layer between legacy ERP software and a centralized production database running PostgreSQL.

Conducted deep-dive technical process audits for high-growth B2B firms, converting loose manual operational bottlenecks into structured, low-code programmatic workflow blueprints. Engineered secure relational storage schemas in MySQL, connecting distributed lead capture infrastructure to centralized databases via authenticated REST APIs.

Designed and implemented custom software automation and operational workflows for various independent clients.

Certification in Designing Agentic Systems with advanced focus on multi-agent architectural design, semantic vector space retrieval, LLM evaluation loops, and orchestration graphs.

Master's degree from Hindu College, University of Delhi, graduating in 2021.