Sravya Thotakuri

About

Hands-on engineer focused on productionizing LLM-powered assistants: builds RAG pipelines with guardrails, response schemas, and citation-grounded outputs, then hardens them with explicit NFRs (latency, uptime, security, cost). Experienced diagnosing agentic/LLM workflow issues in real time using observability and stepwise isolation, and supports go-to-market via developer demos, workshops, and pre-sales technical evaluations in microservices/Spring Boot environments.

Experience

Education

Key Strengths

• Takes LLM prototypes to production by translating assumptions into explicit non-functional requirements (latency, uptime, security, cost)
• Reduces hallucinations via RAG grounding, standardized prompts, response schemas, and citation requirements
• Improves real-world robustness by handling messy user inputs with preprocessing, intent detection, and query normalization/rewriting
• Performance and scalability tuning under load (profiling bottlenecks, caching, load balancing) to control latency spikes
• Real-time troubleshooting of LLM/agentic workflows using observability signals, structured logs, and step-by-step isolation
• Developer-focused technical demos anchored in real code and end-to-end service flows (Spring Boot/microservices)
• Effective pre-sales and adoption partnership with sales/customer teams (integration, security, performance Q&A; tailored demos/workshops)