Developers Adopt Agentic Tools For Production Coding
On 14 March 2026 the author summarized highlights from a February 2026 Pragmatic Summit fireside chat in San Francisco about agentic engineering and coding agents. The discussion covered stages of AI adoption, trust and testing strategies (red–green TDD, manual tests, Showboat), conformance-driven development, codebase patterns, and prompt-injection risks, illustrating practical guidance for developers integrating LLM-driven agents into software workflows.
Key Points
- 1Progression: Developers move from ChatGPT prompts to agents producing most project code within months.
- 2Testing: Red–green TDD and manual curl tests raise reliability and catch issues beyond unit tests.
- 3Tooling: Use templates, conformance suites, and Showboat to ensure consistent quality and audit trails.
Scoring Rationale
Practical, actionable developer guidance increases utility across teams, but limited by single-speaker perspective and anecdotal evidence lacking broad validation.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
