Swarag Reddy Pingili

About

AI/back-end engineer at Canon who helped build and operate an internal production LLM platform that acts as a secure middle layer between users and models, defending against jailbreaks/prompt injection while enabling RAG, memory, and grounded responses over company data. Experienced with LangChain/LangGraph orchestration, vector DB retrieval, and reliability practices (testing, monitoring, adversarial prompts) to run high-throughput, low-latency AI workflows in production.

Experience

Education

Key Strengths

• Built and deployed a production internal LLM platform with a security middle-layer
• Deep practical expertise securing LLM workflows (prompt injection/jailbreak defenses, RBAC, scoped data access)
• Designed and implemented RAG + conversation memory using embeddings and vector databases
• Hands-on orchestration of multi-step LLM workflows using LangChain/LangGraph
• Reliability-focused engineering: unit/integration testing, adversarial prompt testing, failure-mode design
• Data-driven performance evaluation using logging/monitoring and iterative tuning (retrieval parameters, prompts, caching)
• Experience optimizing for high throughput and low latency in retrieval-heavy systems
• Effective collaboration with non-technical product/business stakeholders via demos and outcome-focused translation

Reference Highlights

• Strong analytical skills
• Sound engineering judgment
• Translates abstract requirements into reliable implementations
• Methodical problem-solving approach
• Clear communication with technical and non-technical stakeholders
• Strong sense of ownership
• Focus on real-world constraints (latency, scalability, maintainability)
• Effective cross-functional collaborator
• Aligns technical decisions with product and business requirements
• Proactive about gathering feedback early
• Receptive to input and responsive to concerns
• Explains complex concepts clearly and adjusts detail to audience
• Calm and effective communicator
• Understands productionizing AI beyond model performance (usability/alignment)