This Hacker News discussion revolves around a proposed new workflow for software development, leveraging AI agents, and contrasts it with traditional methods and existing concepts like asynchronous programming and offshoring. Key themes include:
The Nature of Product Ownership and Clarity
There's a strong emphasis on the necessity of clear problem definition and a decisive product owner for any AI-assisted development workflow to succeed. When specifications are ambiguous or the "product owner" is merely a conduit for others' wishes, the process breaks down.
datadrivenangel highlights the dual nature of their experience: "Worked amazingly when it worked. Really stretched things out when the devs misunderstood us or got confused by our lack of clarity and we had to find time for a call..." * mcny states, "I think it can only work if the product owner literally owns the product as in has FULL decision making power about what goes or doesn't go etc. it doesn't work when a product manager is a glorified in between guy, dictating the wishes of the CEO through a game of telephone from the management." * ch4s3 notes this ideal scenario is "fairly rare in my experience." * jt2190 raises the question of the product owner's technical depth and implicit understanding of system complexity: "For example, how technical is this product owner? Are they assumed to 'just know' that they’re asking for an overly complex, expensive technical solution?" * jmull agrees on the importance of clear problem definition, calling it "incredibly useful for software development, period. A 10x factor, all by itself. Yet it happens infrequently, or, at best, in significantly limited ways." They also point out the assumption that "you already know what you want at the start."
The Risk of Skill Atrophy and Tech Debt
A significant concern raised is the potential for developers' coding skills to diminish if they are not actively writing code, and the risk of accumulating "tech debt" when relying heavily on AI.
lelanthran warns, "This works until you get to the point that your actual programming skills atrophy due to lack of use. Face it, the only reason you can do a decent review is because of years of hard won lessons, not because you have years of reading code without writing any." * bpt3 echoes this sentiment: "Why would I choose to slow myself down in the short term and allow my skills to atrophy in the long term (which will also slow me down)?" * lenerdenator expresses concern that AI agents are "a way to create tech debt on a massive scale while not being able to identify it as tech debt." * dboreham draws a parallel to older computing paradigms: "Those of us who worked in hardware, or are old programmers will find this familiar. Chip/board routing jobs that took days to complete. Product build/test jobs that took hours to run."
Redefinition and Confusion Around "Async Programming"
A major point of contention and confusion is the author's use of the term "async programming" to describe a workflow where tasks are delegated and processed without synchronous, real-time interaction. Many users expected a discussion of traditional asynchronous programming concepts in computer science (like async/await in JavaScript).
keybored directly states, "Oh async? > This version of 'async programming' is different from the classic definition. It's about how developers approach building software. Oh async=you wait until it is done. How interesting." * sirwhinesalot feels similarly: "Before I read the article I thought this meant programming with 'async'. Just call it Agent-based programming or somesuch, otherwise it's really confusing!" * drob518 agrees: "Exactly. I think the traditional meaning of 'asynchronous programming' was coined first. So, let’s stick with that." * grandiego also had the wrong impression: "Same here. I've read the author's braintrust.dev as 'brain - Rust - Dev', so I was expecting a discussion on Rust Async development." * lolive elaborates on the difficulty of teaching actual async programming: "Man, for the first time in HN, I am teased to actually read the article. Update: oh my god, I read the article. And feel completely cheated!!!!" and later, "telling them that you should not 'aList.foreach(asyncMethod)', but you'd better do 'Promise.all(aList.map(asyncMethod))' is NOT very easy for them." which highlights the existing complexity of the term they believe the author is reappropriating. * The author, ankrgyl, acknowledges: "Whoops sorry, I can understand how the name is confusing/ambiguous! The use of 'async' here refers to the fact that I'm not synchronously looking at an IDE while writing code all the time."
Alternatively, some users differentiate the author's "async" from "vibe coding" or "AI pair-programming."
didibus attempts to clarify the author's proposed term: "In ??? coding, the developer specifies code changes that must be made... The AI delivers the complete set of changes... and the developer reviews it at the code level." They then suggest "AI pair-programming or AI-assisted coding" as existing terms. * didibus also directly asks ankrgyl to "please DO NOT call it async programming (even if you add async AI it's too confusing)."
The Value of AI in Shifting Developer Roles and Capabilities
Despite the confusion and risks, many users see potential in AI agents to augment development, change how developers interact with code, and handle tedious tasks.
gobdovan suggests a cognitive benefit: "After reading so much AI-generated code with subtle mistakes, I can spot errors much quicker even in human-written code." * curiousC proposes a future where code reviews become rare escalations: "We don't do code reviews on compiler output (unless you're writing a compiler). The way forward is strong static and analytic guardrails... You should rarely have to look at the code, it should be a significant escalation event..." * webstrand finds value in offloading disliked tasks: "I find myself turning to AI to write code I have an aversion to writing... Like benchmarks, bash scripts, dashboards, unit tests, etc." * antimoan frames AI as enabling a return to proper programming stages: "With AI developers are forced to think about the functionality and the specs of their code to pass it to AI to do the job and can no longer just jump to step 3." * datadrivenangel offers a pragmatic view on AI versus lower-tier offshoring: "AI will give you slop code faster and cheaper, and that is sometimes enough." * dec0dedab0de believes AI will shine in "figuring out what you want is the hard part about programming... because it lowers the time between iterations and experiments."
The Philosophical and Experiential Aspects of Coding
Several comments touch on the intrinsic value and enjoyment of the act of coding itself, questioning whether AI-driven workflows detract from this.
suddenlybananas finds the proposed workflow unappealing: "This kind of workflow really doesn't appeal to me in the slightest. Maybe it works for some people, but it just seems to drain all the pleasure out of programming. For me, at least, solving the little problems are like little satisfying puzzles which makes it easier to maintain motivation." * ge96 expresses a similar sentiment: "I actually like writing code, it does get tedious I get that when you're making yet another component. I don't feel joy when you just will a bunch of code into existence with words. It's like actively participating in development when typing." * snozolli sees a future where their least favorite task becomes primary: "The thing I like least about software engineering will now become the primary task. It's a sad future for me, but maybe a great one for some different personality type."
The Role of Visionary vs. Pragmatic Leadership
The discussion also touches upon how business leaders perceive and utilize these new tools, and the potential for both overenthusiasm and shortsightedness.
segfaultex states, "This is what a lot of business leaders miss. The benefits you might gain from LLMs is that you are able to discern good output from bad. Once that's lost, the output of these tools becomes a complete gamble." * bpt3 bluntly adds, "The business leaders already can't discern good from bad."
The Future of Software Development: Skepticism and Optimism
Overall, opinions are split between those who see potential in AI-driven workflows, albeit with significant caveats, and those who are deeply skeptical about their long-term viability or their impact on the craft of programming.
kylereeve predicts a future clean-up: "When this bubble finally pops, someone is going to have to clean up all the nonsense AI code out there." * jmull is direct: "This vision of AI programming is DOA." (though the author ankrgyl later pushes back on this). * datadrivenangel offers a nuanced take on AI's impact on offshoring: "Near term I think that at comparable cost ($100s of dollars per week), it'll give faster results at an acceptable tradeoff in quality for most uses."