What AI Can and Cannot Do Today in Software Development
what AI actually does well in software development, what it can't do yet, and why. The 5 integration levels and the junior developer paradox.
Articles · page 2
The Real Interests of Each Stakeholder in Enterprise AI Adoption
CEO, CTO, developer, manager, and user: what each one really wants from AI, where their interests clash, and what trade-offs each position entails.
AI in Business: Expectations vs. Reality — Who Benefits, and at What Cost?
Article series on what each stakeholder gains and loses from enterprise AI: CEO, developers, managers, and users. Informed opinion, no hype.
CD (Continuous Delivery) — Part 6: Final validations, Manual Approval, and Deploy to Production
Consolidation of all pipeline gates, informed human approval, and progressive deployment to production with Blue/Green or Canary.
CD (Continuous Delivery) — Part 5: LCA — Load, Stress, and Chaos Engineering
Real resilience validation: load tests, stress tests, and chaos engineering to guarantee stability, scalability, and fault tolerance.
CD (Continuous Delivery) — Part 4: Deploy to Staging and Performance Smoke Tests
Pre-production deployment with production parity, E2E and lightweight performance validation, and manual approval before production.
CD (Continuous Delivery) — Part 3: Deploy to Dev and Smoke Tests
Deployment of the verified artifact to the development environment and quick validation with smoke tests to detect obvious failures.
CD (Continuous Delivery) — Part 2: Image Scan and Service Tests
Container image vulnerability scanning and service validation with contract tests, endpoints, and dependency checks.
CD (Continuous Delivery) — Part 1: packaging, SBOM, and remediation
Start of the continuous delivery cycle: Docker packaging, SBOM generation, image signing, and the concept of remediation.
CI (Continuous Integration): analysis, tests, quality, and security
The stages of the continuous integration pipeline: linting, build, deep static analysis, unit and integration tests, service tests, coverage, quality gate, and security scan. With the role of AI in each stage.
Pre-CI: General CI/CD Pipeline Architecture and Pre-checks
High-level view of the complete CI/CD pipeline and the checks that run before CI: fast static analysis, linting, formatting, conventions, contract validation, and AI review.
CI/CD: Branching, Git Flow, and Pull Requests
Branching strategies, GitFlow models, the modern AI-assisted developer workflow, and the complete Pull Request process in professional CI/CD environments.