Dr. Rajat Dandekar helps you turn AI theory into working systems
I’m Dr. Rajat Dandekar, co-founder of Vizuara. I teach engineers to connect AI fundamentals with working code and careful evaluation.
Career highlights
• PhD in Mechanical Engineering, Purdue University; BTech and MTech, IIT Madras.
• Co-author of Manning’s Build a DeepSeek Model (From Scratch).
• Research on complex physical systems, including particles and organisms in stratified flows at Purdue.
• Taught both phases of Vizuara’s RL in Production, from RL foundations to LLM post-training projects.
Vizuara brings first-principles explanations, live coding and research-led teaching to a global audience. Its learner-review collection reports a 4.96/5 average across rated reviews. Explore our teaching on the Vizuara YouTube channel.
Courses
Multi-day, guided programs to get real results



vLLM & SGLang Engineering: End-to-End LLM Serving
Benchmark, tune and deploy both vLLM and SGLang, then choose the right engine using measured latency, throughput and GPU cost.



RL in Production: LLM Post-Training, Coding Agents, World Models and Robots
Understand how RL trains reasoning models, coding agents and robots by building and measuring a small working version of each.



AI Forward Deployed Engineering: From Customer Brief to Deployment
Lead an AI delivery engagement from discovery to deployment, customer acceptance and operational handover.




Kernel Engineering: From CUDA Foundations to Frontier AI Kernels
Write, profile and optimize GPU kernels, then apply modern kernel techniques in a graded engineering capstone.
Lightning Lessons
Free, interactive sessions to explore new topics

DeepSeek from Scratch: V3 to V4.1 Flash

How Trillion-Parameter Models Are Served, From Scratch

From Customer Brief to Working AI Pilot: A Live FDE Build

Inside an LLM Serving Engine: How vLLM and SGLang Work

Inside AI FDE: From Customer Brief to Technical Delivery

How LLM Serving Works: An Introduction to vLLM and SGLang
Resources

Build an AI News Agent with Claude Code
A public Vizuara coding session: build an AI news aggregator with an LLM agent using Claude Code.
Watch the free build

Context Engineering: Foundations of Context
Watch Vizuara’s full introductory lecture on the foundations of context engineering for LLM applications.
Watch the free lecture

MCP: Connect AI Agents to Tools
Learn Model Context Protocol with Dr. Sreedath Panat in this full lecture from Vizuara’s Context Engineering Bootcamp.
Watch the free lecture
Alumni reviews
Tai
vLLM & SGLang Engineering: End-to-End LLM Serving
