minware reposted this
VCs are pouring billions into AI writing code, while engineers spend hours filling out Jira tickets, tracking time, and writing status reports. Instead of replacing engineers, maybe first we should automate tedious non-engineering work that wastes their time.
Wrong end of the power dynamic though. If you automate the non-engineering tasks, you’re a step away from automating middle management. Which is clearly scary and bad (for middle management). Like you, I wish there were more focus here, but I don’t think it’s going to be popular with decision-makers.
We are more than a decade away from replacing people understanding logic. Tickets and tracking are built by logical people for logical people. What you are asking is to spend more time trying to understand what a non logical machine put in there, which will cost more time for the logical people. So no, these aren't the typical tasks that should be automated. Automation is about improving the efficiency of repetitive task. Logical people don't create repetitive tickets. So their uniqueness means that logic prevails.
I agree. I also worked with a wise person who often said "You get what you measure". Often these tasks are imposed on teams (developers, testers) to try and measure their behavior, but rather than objectively getting a measurement, you end up changing behavior in ways you didn't intend. Best example was a place I worked that began counting defect tickets found/submitted by testers. Over time the number of "found" bugs ballooned, which looked great on paper, but in fact one core defect in the code was being written up multiple times (one for iOS, one for Android, one for the web client, etc) when in fact it was one bug that manifested itself in a lot of places. So, in the end, the company got exactly what they measured - but it ended up causing tons of bloat and overhead in tracking/testing/fixing tickets that were, in the end, meaningless. Same can happen on the development side (counting LOC or commits or etc). Ironically, the team knows exactly who is producing quality output and getting stuff done, it's often leadership that imposes this stuff to try and get a handle on how the team is doing, which can, ironically, ruin their productivity. It's the "observer effect"
Back in the day, we had minimum wage/cheaper Admins to do all the tedious tasks of the high-priced assets. Then companies stopped referring to people as assets and gradually dumped all the Admin stuff on said assets. It only took 3-5 years for "ask your Admin to go away". The Admins that survived, mostly became receptionists or executive assistants. Some got entry-level jobs. The bulk left the industry with all that knowledge of the inner workings of their compaies. Ultimately, the move turned a whole lot of assets into liabilities and insider threats. Bringing the "overworked, underpaid and unappreciated" meme from government employees to the private sector.
Spot on! Automating non-engineering tasks would free up engineers to focus on what they do best—solving problems and building great products.
I truly admire your vision for reducing non-engineering tasks to allow engineers to focus on impactful work. If I may, I wanted to ask about something that’s been on my mind. I recently applied for a remote junior frontend position at Minware and received a rejection without an interview. While I understand there may be many reasons for this, I wonder if my location or time zone played a role. I want to emphasize that I’m highly flexible with my working hours and can overlap with any required time zones. In fact, I often work nights in my country, which aligns with the daytime hours of most regions. My motivation for applying was not just about securing a role but being part of Minware's mission to innovate and create meaningful solutions. I hope this message doesn’t come across as unprofessional—it’s simply a testament to how much I resonate with your company’s goals. I’d be grateful for any feedback or guidance that could help me improve for future opportunities. Thank you for your time, and I appreciate your inspiring posts on LinkedIn.
SRTs baby! Simple, Repeatable Tasks are the obvious choice for AI in its current state. Let's not go down the path of writing a paragraph in an LLM for it to spit out 10 lines of code that you, as a developer, should have known how to do in the first place. However, you should know how to ask AI what you want in a way that AI understands and can process. Think of it as a new use of P-Code.
Great point ! Though do you think that the Mythical Man Month https://en.wikipedia.org/wiki/The_Mythical_Man-Month still applies in the era of AI ?
I think, people hated developers in general. It was this high paying, nerdy glamours field, with lots of money. AI just helped to amplify this. It's like shaming an engineering student with a electrician's job. The whole buzz around writing code automatically was a lot more than in places where a word ejaculator and pattern matcher, and low impact fields in the industry was completly overlooked. There is also this whole news around, we don't need junior engineers anymore. I don't know, this kind of statement surely does come from people who lack basic understanding. I mean people don't give birth to senior devs. AI is great, people are the problem.
Lobi Software Studio | The Panic Loop Podcast
1moSo you want machines to perform human tasks while humans write machine code? And how does AI fill out a Jira ticket exactly? What are the requirements, what are the acceptance criteria? What are the considerations that developers want passed to product or vice/versa? Engineers don’t build the bridge. They plan the bridge. They exercise oversight over the build. They adapt as obstacles are encountered and communicate vision to various teams. What you’re referring to as engineers are just code-writers. And AI absolutely can and should replace code-writers.