AI-ERA SOFTWARE DEVELOPMENT · 10 CONNECTED COURSES
Beyond technology names
Why it changed
From the static web to AI Agent operations, connects the constraints of the time → transition mechanism → gains → new costs → current verification criteria in a single flow.
Follow the history from Course 01LEARNING ROADMAP
The 10 courses form a single service lifecycle
Each card is a separate course page. Read them in order, or start with the course you need now.
- 01
Evolution of web development and modern architecture
Connects the reasons web architecture changed, from static documents to dynamic services, SPA, and server rendering, through a single request path.
- 02
How AI has changed development
Learn how permissions and feedback loops change from autocomplete to coding agents that explore, modify, and verify a repository.
- 03
Understanding AI agents
Distinguish chatbots that produce answers from agents that change real state, and design the boundaries of context, memory, tools, policy, and evaluation.
- 04
Connecting AI systems with MCP
Splits the problem of connecting each agent in its own way into host, client, and server contracts, and designs tools, resources, prompts, and security.
- 05
Databases in the age of AI
Study relational constraints, JSON, vector search, and RAG not as substitutes for one another but as answers to different questions with different failure paths.
- 06
Testing and AI development quality
Divide the defects that each layer can detect, from type checks through unit, integration, contract, and browser E2E tests, and close the AI correction loop with release evidence.
- 07
Container and cloud infrastructure
Reduce development-environment differences with images and deployment contracts, and connect Compose, Kubernetes, probes, rollout, and rollback.
- 08
High-performance systems and language choices
Compare TypeScript, Python, Go, and Rust by latency, throughput, memory safety, ecosystem, and cost of change, not by trends.
- 09
Agent operations and observability
Link traces, metrics, and logs to the agent's model, prompt, tool, and policy revisions, and combine cost, quality, and permissions into a release gate.
- 10
AI Native service capstone project
Connect React/Next.js, TypeScript/Python APIs, PostgreSQL/pgvector, Agent·MCP, tests, containers, and observability into a single body of handover evidence.