CS @ San Jose State University • AI Systems • Backend & Data Pipelines
- Build production-style AI systems (RAG, multimodal, local LLMs)
- Design backend systems and high-throughput data pipelines
- Focus on real-world deployment, not just demos
- Co-founded GradAchiever → built automation + data systems for student planning
- Built Java + Python data pipelines for analytics and large-scale data extraction
- Developed a multimodal RAG system (text, image, audio, video) with local LLMs
Languages
Python • Java • C/C++ • SQL
AI / ML
RAG • PyTorch • LangChain • Ollama • LLaVA • Embeddings
Backend & Infra
PostgreSQL • REST APIs • Docker • Data Pipelines
Systems / Tools
Qdrant • Tesseract • FFmpeg • Git
👉 https://github.com/krishangnaikar/Multimodal-Enterprise-RAG
- Built a local, privacy-first AI system (no external APIs)
- Integrated LLMs + vision + embeddings + vector database (Qdrant)
- Supports text, image, audio, and video ingestion + retrieval
- Designed as a modular, production-style ML system
- Built an end-to-end ML pipeline with training and inference
- Implemented transfer learning (VGG16 / DenseNet)
- Designed a reproducible CLI-based workflow for experimentation
- Software Engineering / AI Internships
- Roles in autonomy, ML systems, or backend infrastructure
- Teams building real-world AI systems at scale
- I don’t just use AI — I build full systems around it
- Strong in backend + data + ML integration
- Experience with local LLM stacks (practical + production-focused)