Under the Hood
Deep dives into the architecture, AI systems, and engineering decisions powering MyComputerTutor.
The Brain Behind the Bot: Subject Classification with Small Language Models
How we route every question to the right subject in under 200ms using a cheap classifier — and the prompt design tricks that made it 96% accurate.
Expertise Adaptation: How We Personalize Every Answer
Same question, three different answers — how we represent learner expertise and bake it into every system prompt.
Video on Demand: YouTube Search as a Learning Layer
Text answers are great. A 4-minute video from a master teacher is better. Here's how we wire YouTube into every response.
The Feedback Loop: Building a Self-Improving Tutor
Every 👍/👎 is a training signal. Here's how thousands of ratings become a better system prompt every week.
System Architecture: End-to-End Flow of a Single Question
Follow one question through every box in our backend — classifier, memory, LLM, YouTube, response, feedback, eval.
B2B Multi-Tenancy: Serving Institutions at Scale
How we isolate tenants, customize prompts per institution, and ship a credible enterprise tier on a startup budget.