Course Outline
Foundations of Architecture for AI Systems
- Defining layers, boundaries, and separation of concerns.
- Applying ports and adapters to AI components.
- Establishing quality attribute scenarios for AI features.
- Utilizing architecture decision records as governance tools.
Evolving Existing Applications
- Prioritizing incremental modernization over complete rewrites.
- Positioning AI services alongside existing monolithic structures.
- Applying Strangler and anti-corruption layer patterns.
- Creating integration seams for legacy APIs and data.
Architectural Patterns for LLM Integration
- Establishing model gateways as abstraction boundaries.
- Managing synchronous, asynchronous, and streaming integrations.
- Implementing fallbacks, degradation, and provider failover strategies.
- 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.
- Managing latency budgets and timeout orchestration.
- Treating cost per request as a critical architectural constraint.
End-to-End RAG Architecture
- Constructing ingestion, chunking, and embedding pipelines.
- Designing indexing, retrieval, reranking, and generation stages.
- Positioning retrieval relative to application services.
- Analyzing trade-offs and potential failure modes at each stage.
Vector Databases and Retrieval Design
- Comparing managed versus self-hosted vector stores.
- Implementing dense, sparse, and hybrid search techniques.
- Applying metadata filtering and tenant isolation.
- Developing re-embedding strategies for model changes.
Retrieval Quality, Grounding, and Evaluation
- Building evaluation sets from real-world queries.
- Measuring retrieval and answer quality metrics.
- Ensuring citation, grounding, and containment of hallucinations.
- Integrating feedback loops back into the index.
Agent Architectures
- Understanding the agent loop and tool invocation mechanics.
- Distinguishing between deterministic workflows and autonomous planners.
- Identifying scenarios where agents are inappropriate.
- Defining agent responsibilities and state management.
Multi-Agent Systems and Orchestration
- Implementing supervisor and orchestrator topologies.
- Defining handoff protocols and message contracts.
- Managing shared memory and context.
- Comparing frameworks based on architectural merit.
Service Communication and Tool Protocols
- Coordinating via event-driven and queue-based systems.
- Exposing tools through standard protocols.
- Managing sagas, compensation, and long-running transactions.
- Ensuring observability across multiple service hops.
AI Agents in the Development Workflow
- Incorporating coding agents into the delivery pipeline.
- Utilizing spec-driven development and review gates.
- Generating tests and running automated quality checks.
- Maintaining architectural control through guardrails.
Governance and Software Quality
- Conducting design reviews for AI features.
- Managing model and prompt changes.
- Integrating regression evaluation into the pipeline.
- Defining ownership, policies, and approval paths.
Security, Privacy, and Data Governance
- Mitigating prompt injection and implementing defense in depth.
- Managing secrets, access control, and least privilege.
- Addressing data residency, personal data, and retention.
- Ensuring auditability and human-in-the-loop controls.
Scalability, Cost, and Observability
- Tracing processes across agent chains.
- Monitoring token usage and cost telemetry.
- Optimizing caching and latency.
- Establishing service levels and incident response for AI services.
Capstone: Architecture for an AI-Driven Solution
- Developing context, component, and deployment views.
- Designing retrieval and agent orchestration.
- Planning for security, governance, and costs.
- Conducting peer reviews, critiques, and migration roadmapping.
Requirements
- Proven experience in designing and sustaining production-grade software systems.
- Proficiency with REST APIs and service integration practices.
- Understanding of relational databases and data modeling concepts.
- Initial exposure to LLM-based tools or APIs.
Target Audience
- Software Architects.
- Technical Leads and Senior Engineers.
- Engineering Managers accountable for architectural 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