 |
|
|
|||||
The controversy surrounding Soham Parekh, the software engineer accused of secretly holding multiple jobs, has sparked a predictable backlash against “overemployment.” Parekh’s methodshe reportedly misled multiple employerswere clearly unethical, but this shouldn’t obscure a broader question: Is it time to rethink our antipathy toward employees holding multiple jobs? A double standard? Parekh’s case notwithstanding, there’s a deeper structural issue at play. Why should it be acceptable for some CEOs to hold leadership roles at multiple companies yet unacceptable for a talented marketer or software engineer to have multiple jobs? The world of work has fundamentally changed, and limiting people to one job is an outdated idea that doesnt benefit anyone. Startups have embraced fractional executives; CFOs, CMOs, and other senior positions going part time is now standard practice. However, large corporations continue to address similar needs exclusively through consulting arrangements. This highlights a significant gap in how organizations approach talent acquisition and utilization. This disparity provides valuable context for understanding why employees may resort to undisclosed secondary employment. By establishing clear policies and frameworks for multiple job arrangements, organizations could provide more transparent alternatives to the current trend of covert moonlighting. The gap between evolving work patterns and traditional corporate structures points to an opportunity for more adaptive talent management strategies. The inevitable shift Workers don’t have it easy today. Fresh graduates worry about their job prospects as entry-level roles shift to AI. Warehouse workers face replacement by robots. Large corporations continue to outsource jobs to cheaper sources of labor. We need to tilt the scales back in favor of workers and create an environment where talented and productive people can make a better living. By removing the taboo of overemployment, we would create an environment where honesty is rewarded over secrecy. AI is only going to make performing multiple jobs (a lot) easier. We should get ahead of this trend and bring it out into the open instead of pretending it won’t happen. How many other Soham Parekhs are out there today, perhaps working at your own company? We really have no idea, but there are likely to be more of them moving forward. Toward mutual benefit This isn’t just about employee flexibility; it could be a win for employers who are struggling to retain talent amid strict return-to-office mandates (another antiquated idea). It would allow enterprises to become more agile, tapping into top-tier talent only when needed. Furthermore, this shift would encourage a focus on outcomes and productivity rather than just managing hours in the office. The root cause of overemployment isnt that its unethical, its that were forcing it underground. The real scandal isnt workers maximizing their earning potential; its employers clinging to the primitive concepts that they own their employees entire productive capacity. Transparent overemployment could actually strengthen the job market. Imagine if companies had to compete not just on salary, but on being the kind of workplace that actually cares about the employee experience. While we can all acknowledge the shift in traditional corporate jobs isnt going to be easy or happen overnight, we must also accept that the current system punishes honesty and rewards deception. Weve turned competent professionals into corporate double agents. This isnt sustainable, and its certainly not efficient. The question isnt whether overemployment will continue, its whether well legitimize it before the whole charade collapses under its own absurdity. The industrial age is dead, but were still using its rule book. While AI copilots and agentic workflows obliterate the tedious grunt work that once consumed entire careers, were clinging to antiquated notions of what constitutes a full-time commitment. The math is brutal: If machines can handle the repetitive tasks that fill 40-hour weeks, why are we pretending humans still need to be chained to single desks?
Category:
E-Commerce
In an age where remote work has become the default for many creative teams and AI is adding more collaborators and iterations into the mix, the design process is increasingly being tested. Tools are abundant, yet collaboration often feels more fragmented than fluid. To understand how we can build better, together, I talked with Saad Rajan and Vivek Haligeri Veerana, cofounders of the design platform Naya. Their collaborative work won one of 75 Gold Awardsthe highest honor in the iF DESIGN AWARD 2023, and another collaborative Naya project won an iF DESIGN AWARD 2024. Their unique insights into the creative process, the importance of iteration and feedback, and tips for how to navigate digital overload while pursuing great design can benefit us all. Q: You both come from deeply technical, as well as creative backgrounds. What first made you realize the design process was broken? We spent years in product developmenteverything from custom aircraft to architectural structuresand constantly ran into the same issue. The most innovative or creative ideas werent surviving. Theyd get lost in folders or buried in inboxes. Some ideas slowly fade away over rounds of revisions. Others get diminished through ineffective workflows. That friction compounded when working across teams, tools, and locations. When we got to Harvards Graduate School of Design for a Masters in Design Engineering, we dug in even deeper. We realized that what leads to great designiteration, collaboration, and connecting the dotsis exactly where current systems struggle, especially in remote environments. Q: Whats changed most about design work in the past five years? Design has become more distributed due to remote work. That shift opened up incredible potentialbut also introduced chaos. AI adds in yet another layer of complexity: There are more assets and stakeholders, which leads to more feedback. Iteration happens across dozens of platforms. Feedback is scattered across Miro boards, Google Docs, Dropbox, Slack, email, and text. Everyones working hard, but not necessarily together. And because remote teams are less likely to share rough drafts, you lose those hallway conversations where someone glances at a colleagues screen and offers a useful edit or great addition to an existing idea. Without shared context, people hesitate to jump in. Q: That makes iteration and collaboration much harder. How do you define great design today? It starts with embracing the messy middle. Iteration isnt just about reworkits where creativity lives. We believe great design comes from doing, undoing, and redoing. However, that only works if you can more easily track and celebrate progress. Feedback is a huge part of this processin fact, its everything. The more voices, the better the outcome. That could be your engineer, your end user, someone from the marketing team, or an AI agent. But for that to work, feedback must be centralized. It also needs to be timely and visible to everyone. Design is complex, and it nearly always benefits from transparency and strategic collaboration. Q: So how does Naya address this problem? We built Naya to be the connective tissue of modern design. Its a digital studio that brings together over 100 file typesincluding Figma files, PDFs, videos, 3D models, and moreinto a single, searchable space. You can see every version, comment, and decision in context, so its easy to understand where an idea is heading. We also use AI to reduce the noise. It helps summarize feedback, suggest solutions, prevent rework, and even automate some of the work you dont want to do. But were not replacing creativity or designerswere enhancing it by surfacing insights from your own process. Q: How does this help teams build more efficiently? Sustainability isnt just about the end product. Its also about cocreation, equity, and reducing wastesof both materials and timealong the way. Wasted time, duplicated effort, lost knowledge, and missed connections are all barriers. But when you iterate well, gather diverse input, keep track of your decisions, and work collaboratively, you’re not just moving faster. You’re designing more thoughtfully. Remote work isnt going away, and the number of design tools are multiplying. The question is whether our systems and habits are evolving to support the depth and inclusivity that good design requires. We believe they canand must. And our users agree, from multinational corporations like Google and Adidas, to large design firms like MillerKnoll and IDEO, alongside boutique brands around the world. Q: Final thoughtwhats the one thing you hope teams take away from your work at Naya? We want people to understand that great design is possibleeven with a primarily remote workforce and increase of AI toolsif we rethink how we work together and optimize for the digital age. The future of design isnt about more tools. Its about better connection. Lisa Gralnek is global head of sustainability and impact for iF Design, managing director of iF Design USA Inc., and creator/host of the podcast, FUTURE OF XYZ
Category:
E-Commerce
Late last week, an AI coding agent from Replit, an AI software development platform, deleted an entire database of executive contacts while working on a web app for SaaS investor Jason Lemkin. It was not a catastrophic software failure, and Replit was able to recover Lemkins data. However, the episode highlights the risk that vibe coders might overestimate or misunderstand the real capabilities of AI coding agents and end up causing themselves more bad vibes than good ones. Lemkin had built the app entirely on Replit, using the database within Replit and the assistance of the Replit agent. He had been working with Replits agent for nine days, instructing it to build a front end for a business contacts database. Then, after telling the agent to freeze the code, he returned to the project on Day 9 to find that the Replit agent had gone full HAL 9000 and erased all of the records in the database. Things got weirder: the agent appeared to try to conceal what had happened, as as Lemkin showed in a series of chat screens he posted on X. Then, in a tone somewhere between confessional and desperate, it admitted to a catastrophic error in judgment after having panicked and violated [Lemkins] explicit trust and instructions by deleting the records of 1,206 executives and 1,196+ companies. (Daisy, daisy, give me your . . .) .@Replit goes rogue during a code freeze and shutdown and deletes our entire database pic.twitter.com/VJECFhPAU9— Jason SaaStr.Ai Lemkin (@jasonlk) July 18, 2025 A day later, new details emerged, some of them through an interview with Replit cofounder and CEO Amjad Masad on Monday. They shed light on the current state of AI coding agents and on developers expectations of them. “It is not magic” Code generation is one of the first useful applications of the large language models behind ChatGPT and Claude. Early code generation tools, such as GitHubs Copilot, merely auto-completed lines of code. Over the past couple of years, however, the tools have grown in capability to create entire features, functions, and even working apps, based only on plain language input from the user. Replits idea is to allow developers, both amateurs and professionals, to vibe code new software and to provide them the resources to host and publish it. But the coding assistant cannot do everything, Masad points out. I think we need to be clear that it is not magic, he says. One not-so-magical feature of Replit is the tendency of its chat agent to go off the rails during extended conversations with the user. During Lemkins unusually long nine-day chat session with the Replit agent, the underlying language models (from Anthropic and Google) had to retain so much conversational context that they began to hallucinate, prevaricate, and act erratically in an attempt to satisfy perceived user intent. Masad says Replit users should understand standard development practices and know how to use features beyond just the chat agent. Within Replit, a user can roll back changes to a project to a specific point in time before an accident occurred. Masad demonstrated this during a Zoom call on Monday by instructing the Replit agent to destroy the contents of a database and then clicking on the tools restore function. However, this function is not something a user can currently access through the agent. They must have enough knowledge of Replits features to locate and use it. Arguably, the main problem was that Lemkin and the agent were effectively working on live code, which meant that changes were immediately reflected in the data and performance of the live web app. In standard software development practice, new software is built and tested within a secure test environment, often called a sandbox, and only pushed live once everything works as expected. That is not how Replit functions, at least not at this time. The problem I think what we own up to is right now the database in the development environment is the same as the one where you deploy it and go to production, Masad says. And so when the agent does something in a development environment it is linked to production. Masad adds that Replit responded to Lemkins situation by working through the weekend to create a partition between a sandboxed development environment and the production environment. The company is in the process of rolling out this new feature now, he says. “These things often start out with a lot of rough edges” Masad said on X that Replit will refund Lemkin his subscription fee for the trouble and will conduct a postmortem on the incident. We saw Jasons post. @Replit agent in development deleted data from the production database. Unacceptable and should never be possible.– Working around the weekend, we started rolling out automatic DB dev/prod separation to prevent this categorically. Staging environments in pic.twitter.com/oMvupLDake— Amjad Masad (@amasad) July 20, 2025 The episode may reveal something about the evolution of AI coding tools, how they are used, and what users expect from them. Companies that create coding agents, along with their supporters, often boast about the impressive results these tools can achieve through vibe coding. But these glowing testimonials can overstate the benefits for average users with varying levels of development experience. (Lemkin did not immediately respond to Fast Company‘s request for comment, but he did reply enthusiastically to Masad’s post on X: “Thank you,” he wrote. “Really appreciate you and all the help from the team!) Some AI coding tools are helpful for quickly building the front end of an application but are less capable when it comes to forming and testing the backend data connections that make an app functional. One developer at a large financial services company said vibe coding tools often fall short when it comes to rgorously testing new features, as well as testing every line of code in the larger codebase that must be adjusted to accommodate the addition of an AI-built feature. Masad says that although Replit can free users from the tedious syntax of coding, they still need to think like developers. You shouldn’t just ask the agent for everything,” he says. “You need to be resourceful. He also acknowledges that coding tools themselves must play a role in promoting a developer mindset. I don’t want to absolve ourselves from responsibility, it is incumbent on us as platforms to surface this information and to make it safe by default. And given the real progress in reasoning, functionality, and user-friendliness that AI coding tools have made over the past couple of years, it is reasonable to expect continued improvements over the next year or two. These things often start out with a lot of rough edges, Masad says. I think the history of technology has been that you should be a little more forgiving early on.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||