Async Programming as a Workflow: Specify, Automate, Review

Async programming is reframed as a workflow where developers write precise specs, delegate implementation to agents, and review results later. It relies on clear problem definitions, automated verification in CI, and careful human code review. This shift lets teams parallelize work and focus on architecture and quality rather than typing.

Key Points

• Async programming here is a workflow shift: write precise specs, delegate implementation, and review later, rather than coding synchronously.
• Success depends on three pillars: clear problem definitions, automated verification (tests, types, benchmarks, linting), and thorough human code review.
• Automated checks—ideally in CI—let agents validate work independently, reducing manual edge-case testing.
• Developers can parallelize work, handling multiple background tasks while focusing on planning, architecture, and review.
• Braintrust applies this approach and builds tools like Loop to automate evaluation and iteration on AI systems.

Sentiment

The Hacker News community is predominantly skeptical of the article's thesis. While a minority share positive practical experiences with AI-delegated workflows, the dominant voices express concerns about the confusing terminology, the false assumption that problem definition is easy, skill atrophy risks, and tech debt accumulation. Many see this as a repackaging of existing management patterns rather than a genuine innovation. The discussion is thoughtful rather than hostile, with the author engaging constructively, but the overall lean is that this workflow has significant limitations the article glosses over.

In Agreement

• Developers on parental leave or with limited time find this specify-delegate-review workflow genuinely productive, enabling them to maintain code quality through familiar review processes while working in short bursts
• AI delegation works well for boilerplate and non-novel tasks like Terraform, dashboards, and test scaffolding — tasks developers understand but find tedious to write manually
• The distinction from vibe coding is meaningful: specifying at the code level with rigorous review is fundamentally different from just prompting for features and accepting whatever output comes back
• AI dramatically lowers the cost of experimentation and exploration, making it easier to try approaches before committing to a specification
• Multi-agent review pipelines with static analysis guardrails could eventually reduce the need for manual code review, similar to how we don't review compiler output

Opposed

• The term 'async programming' hijacks an established computer science concept, causing confusion and undermining the article's credibility
• Clearly defining problems upfront is actually the hardest part of software development — the article treats it as straightforward when it rarely is in practice
• Skill atrophy is a serious concern: developers who stop writing code will gradually lose the ability to effectively review it, creating a dangerous feedback loop
• This workflow is essentially what tech leads and product owners already do with human teams — it's not a new paradigm, just the same management pattern with AI replacing junior developers
• AI-generated code creates tech debt at massive scale while making it harder to identify as tech debt, and once developers lose the ability to discern good from bad output, the quality becomes a gamble
• Reviewing code you didn't write takes longer than writing it yourself, making this workflow slower in practice despite claims of parallelism
• This approach drains the pleasure from programming — solving problems is inherently satisfying, and delegating that work removes the primary motivation for many developers