Navdeep Singh

About Navdeep Singh

AI/ML Engineer specializing in large language models (LLMs), multi-agent systems, and scalable deep learning infrastructure for real-time applications. I am passionate about designing, building, and deploying production-ready AI systems that deliver measurable business value.Key accomplishments include- Leading the development of Agent Charlie, a multi-agent conversational AI with Microsoft AutoGen, reducing token usage by 40% and enabling efficient real-time decision-making for agriculture and farmer assistance- Architecting transformer-based Q&A systems that outperform GPT-3.5 baselines by 18%, leveraging LoRA, quantization, and ONNX Runtime for scalable, low-latency deployment- Delivering high-accuracy (85%+ top-3) recommendation systems and building automated pipelines that improve efficiencies by up to 70%.My expertise spans LLM integration, agent orchestration, transformer models, inference optimization, and ML deployment on cloud infrastructure (Azure). I enjoy working at the intersection of deep learning, scalable software, and practical real-world impact.I am actively seeking opportunities to bring advanced ML research and engineering into production—especially where innovation in LLM and agent systems can transform businesses.

Experience

1. Ai Ml Engineer Intern

   Undone

   Jan 2026 — Present · San Bruno, CA, US

   Architected and deployed an AI-powered recommendation engine on Google Cloud Platform (GCP), leveraging machine learning algorithms and cloud-native services (Cloud Run, Cloud Functions, Firestore) to intelligently prioritize and sort student homework assignments based on urgency, difficulty, and learning objectives — improving task completion efficiency and time management for K-12 users- Engineered LLM-powered content processing pipelines using large language models (GPT, prompt engineering, NLP) deployed on GCP infrastructure, dynamically transforming and simplifying academic material — making complex study content more accessible, engaging, and age-appropriate for young learners- Leading the end-to-end AI/ML ecosystem strategy and cloud architecture for an EdTech student platform, driving the integration of generative AI, natural language processing (NLP), recommendation systems, and intelligent automation on Google Cloud Platform (GCP) — leveraging Cloud Functions, Firestore, Cloud Storage, CI/CD pipelines, and containerized microservices (Docker, Kubernetes) to deliver scalable, personalized, adaptive learning experiences that educate and empower students at scale.

Education

• California State University - East Bay

  Master of Science - MS, Computer Science

• Bikaner Technical University | University College Of Engineering & Technology

  Bachelor of Technology - BTech, Computer Science