Full-Stack SWE - Distributed Systems & AI Pipelines

Di Pan

Houston, TX 713-***-**** ***********@*****.*** linkedin.com/in/dipan121 github.com/Rolinmuuu SUMMARY

Rice University MS Statistics candidate seeking a Summer 2026 SWE Internship or Jan 2027 Full-Time role. Blends rigorous analytical skills with hands-on expertise in full-stack development, cloud architecture, and AI integrations. Specialized in architecting high-concurrency distributed systems (Java, Go) and building intelligent data pipelines

(RAG, LLMs) to translate complex data into scalable engineering solutions. EDUCATION

Rice University Houston, TX

Master of Science in Statistics Aug 2025 - Dec 2026 Southeast University Nanjing, China

Bachelor of Art in Finance, GPA: 3.9/4.0 Sep 2021 - Jun 2025 TECHNICAL SKILLS

Programming Languages: Java, Python, Go, Kotlin, JavaScript/TypeScript, SQL, R Frameworks & Backend: Spring Boot, React, Express.js, Node.js, Redis, CI/CD, Microservice, Kalfa, ELK stack Cloud, Databases & Tools: AWS (ECR, RDS), GCP, Docker, Git, PostgreSQL, MySQL, Elasticsearch, Prometheus AI & Statistics: LangChain, RAG, AI Agent Prompt, Vibe Coding, PyTorch, TensorFlow, Pandas, A/B Testing PROJECTS

OnlineOrder: High-Concurrency E-Commerce Platform (Spring Boot, Redis, AWS)

● Designed and implemented a modular Spring Boot MVC backend, developing 20+ RESTful APIs supporting user registration, menu browsing, and transactional order processing with optimistic locking and concurrency control to handle simultaneous checkout requests.

● Architected the data persistence layer using Spring Data JDBC and PostgreSQL (AWS RDS), implementing query indexing and Redis caching to reduce average menu query latency by 40% under load testing.

● Built a React + Ant Design Single Page Application (SPA) with centralized client-side state management, enabling real-time shopping cart synchronization and dynamic checkout workflows across multiple user sessions.

● Implemented automated testing pipelines including unit tests and API integration tests (JUnit / Postman collections), achieving 85% test coverage across core business logic.

● Containerized the application with Docker and managed images via AWS ECR, deploying the production build to AWS App Runner with automatic scaling support.

Agent AI: Full-Stack RAG & MCP-Driven Semantic Search Q&A Agent

● Crafted a voice-enabled frontend using React and Ant Design, integrating speech-tts libraries. Optimized performance to achieve sub-500ms response latency for seamless, real-time conversational user interactions.

● Engineered a sophisticated RAG pipeline utilizing LangChain and GPT-5. Enabled deep semantic document analysis and high-accuracy response generation by grounding the LLM in complex, domain-specific data contexts.

● Architected a hybrid search strategy via a Model Context Protocol (MCP) server. Orchestrated data retrieval by synthesizing RAG outputs with real-time SerpAPI web results for comprehensive, up-to-date query responses.

● Built a scalable Node.js and Express backend to manage high-throughput document ingestion. Optimized processing logic to handle 50+ concurrent multi-page document queries while maintaining high system reliability. SocialAI: Distributed AI-Driven Social Network (Go, Kafka, Docker)

● Architected a high-concurrency Go microservices platform using React, Redis, and Elasticsearch. Implemented core social features, real-time messaging, and secure JWT-based authentication for seamless session management.

● Integrated OpenAI DALL-E 3 and Embeddings for AI-generated media and personalized recommendations. Replaced rule-based browsing with vector-based similarity search to enable advanced semantic content discovery.

● Decoupled heavy workloads like feed materialization and notifications using Kafka/Pub-Sub. Optimized system resilience and throughput, ensuring reliable processing of asynchronous tasks under spiky write traffic.

● Containerized services with Docker and built CI/CD pipelines for automated deployment. Integrated unit, integration, and contract tests to minimize regressions and ensure consistent delivery across multiple environments.

● Enhanced reliability via rate limiting and fault-tolerant communication, achieving a 99.9% success rate. Secured the platform with input validation and protected media access to ensure production-grade robustness.

Contact this candidate