Autonomous Driving and Planning
End-to-end planning, latent planning states, perception-to-control pipelines, and controller design for self-driving systems.
AI Researcher · KAIST · AXE Lab
AI Researcher | Integrated MS-PhD Student at KAIST | AXE Lab
I work on artificial intelligence systems that connect theory, autonomy, and real-world deployment, with research interests across reinforcement learning, autonomous driving, planning and perception, natural language processing, large language models, optimization, and interpretable AI. I also contribute to research-to-practice initiatives through QuanSkill, QuanVerse, QuanSolutions and QuanRobotics, connecting AI education, robotics, physical AI, and real-world system building.
About
I am an AI engineer and researcher with experience across academic research, applied AI development, and research-driven product engineering. At KAIST, my current research direction focuses on end-to-end planning for autonomous driving, latent states of planning modules, and control architectures for self-driving systems.
My broader research agenda is centered on building intelligent systems that are not only accurate, but also interpretable, stable, safe, and useful in real environments. This includes reinforcement learning, semantic dynamics in language models, agentic AI, computer vision, robotics, and AI systems for education, healthcare, cybersecurity, and mobility.
Research
My work connects autonomy, representation learning, dynamical systems, and safety-aware AI.
End-to-end planning, latent planning states, perception-to-control pipelines, and controller design for self-driving systems.
Safe and explainable RL, constrained decision-making, symbolic oversight, grammar-constrained generation, and agent safety.
Semantic drift, entropy signals, Koopman operators, spectral methods, hallucination detection, and alignment under domain shift.
Object detection, measurement systems, human pose debugging, robotics perception, IoT security, and applied AI systems.
Research-to-Practice
Beyond academic research, I am involved in building initiatives that help students, engineers, and early-stage researchers move from theory to practical AI system development.
QuanSkill focuses on applied AI education and skill development, including large language models, agentic AI, retrieval-augmented generation, local AI systems, automation, and project-based AI learning.
QuanVerse is an AI learning and community ecosystem designed to help learners explore modern AI, build real projects, understand research ideas, and connect technical learning with practical product development.
QuanSolutions focuses on applied AI solution development for real-world business and industry use cases, including LLM integration, AI agents, workflow automation, RAG-based systems, and client-facing intelligent applications. It reflects my interest in translating AI research into deployable systems with practical impact.
QuanRobotics connects robotics, physical AI, perception, planning, control, simulation, and autonomous systems education. It aligns closely with my research interests in autonomous driving, reinforcement learning, and embodied intelligence.
These initiatives reflect my broader goal of connecting academic research, AI education, and real-world deployment through structured programs, applied systems, and hands-on technical mentorship.
Publications
Scholar statistics and publication metadata are based on the profile snapshot provided by Satyam Mishra.
Projects
Developed and presented a healthcare AI agent PoC focused on domain-specific knowledge bases and agentic AI frameworks.
Built a research-grade AI agent using Gemini, LangChain, tool calling, DuckDuckGo search, and Pydantic-based structured outputs.
Developed a self-driving robot car with deep learning for collision avoidance, path following, and image recognition.
Built a real-time object detection and dimension measurement system using computer vision, reaching over 98% reported accuracy.
Created a Streamlit chatbot using Mistral-7B-v0.1, with practical deployment considerations for generative AI applications.
Built a platform for analyzing email sentiment with Hugging Face Transformers, NLTK, and language-model-based response generation.
Patents
Books
Recognition
Vietnam National University, Hanoi.
13th Student Research Conference, Vietnam National University, Hanoi.
5th International Science Forum 2021, HCMC Youth Forum, Vietnam.
Serving as AAAI 2026 Program Committee member and NeurIPS 2026 reviewer.
Contact
For research collaboration, academic opportunities, and AI systems work, please contact me by email or through my academic profiles.