← Study

AI + Backend

Notes on systems, models, and the backend behind them.

API Design (REST vs RPC vs gRPC vs GraphQL)

Synchronous vs Asynchronous Processing

Caching Strategies

Load Balancing

Rate Limiting & Throttling

Authentication & Authorization

CAP Theorem & Consistency Models

Database Replication & Sharding

Message Queues & Event Streaming

Idempotency & Exactly-Once Semantics

Consensus & Coordination

Container Orchestration (Kubernetes basics)

Relational vs NoSQL for AI Workloads

Object Storage & Data Lakes

Data Pipelines & ETL for ML

Vector Databases

Approximate Nearest Neighbor Search

Feature Stores

Data & Model Versioning / Lineage

Model Serving Architectures

Model Serving Frameworks

GPU/Hardware Utilization Basics

Batching & Dynamic Batching

Model Quantization & Compression

Multi-Model & Multi-Tenant Serving

Autoscaling for Inference Workloads

LLM Inference Fundamentals

Prompt Engineering & Prompt Management

Chunking & Embedding Strategies

Retrieval-Augmented Generation (RAG)

RAG Retrieval: Similarity Search vs MMR

LLM Orchestration Frameworks

Streaming Responses (SSE/WebSockets)

Semantic Caching for LLMs

Fine-Tuning vs Prompting vs RAG

Agentic Systems & Tool Use

LLM Gateway & Multi-Provider Routing

Guardrails, Content Moderation & Safety

ML Model Lifecycle & CI/CD

Experiment Tracking

A/B Testing & Canary Deployments

Model Monitoring & Drift Detection

Cost Optimization for AI Systems

Observability for AI Systems

Fault Tolerance & Circuit Breakers

Capacity Planning & SLOs

Design a RAG-Based Support/Q&A Chatbot

Design an LLM API Gateway / Multi-Tenant AI Platform

Design a Real-Time Recommendation System