Course Outline
Foundational Architecture for AI Systems
- Defining layers, boundaries, and the separation of concerns.
- Applying ports and adapters patterns to AI components.
- Establishing quality attribute scenarios for AI features.
- Utilizing architecture decision records as a governance mechanism.
Modernizing Existing Applications
- Pursuing incremental modernization rather than full rewrites.
- Positioning AI services alongside existing monolithic structures.
- Employing strangler and anti-corruption layer patterns.
- Identifying integration points for current APIs and data streams.
Patterns for LLM Integration Architecture
- Using the model gateway as an abstraction boundary.
- Managing synchronous, asynchronous, and streaming integrations.
- Implementing fallbacks, degradation strategies, and provider failover.
- Treating prompt and model configurations as version-controlled assets.
Contracts, APIs, and Data Exchange
- Defining structured output schemas and validation rules.
- Handling idempotency, retries, and rate limiting.
- Setting latency budgets and orchestrating timeouts.
- Treating cost per request as a key architectural constraint.
End-to-End RAG Architecture
- Constructing ingestion, chunking, and embedding pipelines.
- Orchestrating index, retrieval, reranking, and generation stages.
- Determining the placement of retrieval relative to application services.
- Assessing trade-offs and potential failure modes at each stage.
Vector Databases and Retrieval Design
- Evaluating managed versus self-hosted vector stores.
- Implementing dense, sparse, and hybrid search techniques.
- Applying metadata filtering and tenant isolation.
- Planning re-embedding strategies when models evolve.
Retrieval Quality, Grounding, and Evaluation
- Creating evaluation sets based on real-world queries.
- Defining metrics for retrieval and answer quality.
- Ensuring citation accuracy, grounding, and containment of hallucinations.
- Establishing feedback loops that inform the index.
Agent Architectures
- Understanding the agent loop and tool invocation processes.
- Distinguishing between deterministic workflows and autonomous planners.
- Recognizing scenarios where an agent is an inappropriate solution.
- Defining the scope of agent responsibilities and state management.
Multi-Agent Systems and Orchestration
- Designing supervisor and orchestrator topologies.
- Establishing handoff protocols and message contracts.
- Managing shared memory and context across agents.
- Comparing frameworks based on architectural merit.
Service Communication and Tool Protocols
- Utilizing event-driven and queue-based coordination.
- Exposing tools through standard protocols.
- Implementing sagas, compensation logic, and long-running transactions.
- Ensuring observability across multiple service interactions.
AI Agents in the Development Workflow
- Integrating coding agents within the delivery pipeline.
- Adopting spec-driven development with review gates.
- Generating tests and automated quality checks.
- Maintaining guardrails that preserve architectural control.
Governance and Software Quality
- Conducting design reviews for AI features.
- Managing changes to models and prompts.
- Incorporating regression evaluation as a pipeline stage.
- Defining ownership, policies, and approval pathways.
Security, Privacy, and Data Governance
- Addressing prompt injection and implementing defense in depth.
- Managing secrets, access control, and least privilege principles.
- Ensuring data residency, personal data handling, and retention compliance.
- Maintaining auditability and human-in-the-loop controls.
Scalability, Cost, and Observability
- Tracing performance across agent chains.
- Monitoring token usage and cost telemetry.
- Optimizing caching and latency.
- Defining service levels and incident response for AI services.
Capstone: Architecture for an AI-Powered Solution
- Developing context, component, and deployment views.
- Designing retrieval and agent orchestration strategies.
- Creating plans for security, governance, and cost management.
- Engaging in peer review, critique, and migration roadmap planning.
Requirements
- Experience in designing and maintaining production-grade software systems.
- Proficiency in REST APIs and service integration.
- Knowledge of relational databases and data modeling techniques.
- Basic familiarity with LLM-based tools or APIs.
Target Audience
- Software architects.
- Technical leads and senior engineers.
- Engineering managers accountable for architecture decisions.
Testimonials (1)
That i gained a knowledge regarding streamlit library from python and for sure i'll try to use it to improve applications in my team which are made in R shiny