Lennox Prince

Jenny Wen from Anthropic was on Lenny's podcast recently talking about how design is changing & I've seen it shared a lot with very little pushback and lots of hyperbole so I want to add some balance. There are 3 things in here that are genuinely worth sitting with. 1) She puts numbers on the time allocation shift - she's gone from 60-70% of her time mocking & prototyping down to 30-40%, with the rest now spent pairing with engineers & shipping code directly. That's real data from her context & it marries with anecdotal evidence I see from others who've found genuine efficiencies in their personal workflow - but what you do with that reclaimed time is going to vary dramatically depending on where you sit, what you're building & who you're building it for. 2) She talks about vision horizons compressing - no more 2 or 5 year design visions at Anthropic, they're working 3 to 6 months out & the output is a prototype that points a direction, not a polished deck. 3) Her three hiring archetypes super interesting - the block-shaped generalist who's 80th percentile across multiple skills, the deep T specialist who's top 10% in 1 thing, & the craft new grad who learns fast & isn't weighed down by old process. That last one I'm a huge advocate of - the value of a graduate with strong foundational skills who hasn't yet been told how things are supposed to work is massively underestimated. However I think the conversation around this episode has been a bit uncritical. Jenny says the traditional design process is dead. But when it comes to what replaces it the answer is essentially... be flexible, pair with engineers, set short-horizon vision. And in my opinion she hasn't offered an alternative process with any real detail. I don't think she's claiming to, but the way people are sharing this you'd think someone just handed us the new playbook. She didn't. That vision compression, that faster prototyping all only works if your entire company is oriented to working this way and again, lets not fetishise the single domain product design/dev challenge. I think it's important to say - this is one designer's perspective from inside one of the fastest moving AI labs on the planet. If you're a design lead at a bank or an insurer or a healthcare company, where you've got regulatory constraints, slower release cycles and engineers who aren't running seven Claude instances... there is nothing in this episode that is going to dramatically change your Monday morning. The frontier is useful context. It is not a universal strategy. Take the main points but don't fall for the 'XYZ is dead' or 'XYZ role is no longer needed' The hard work of figuring out what the new process actually looks like for your team, your constraints, your product - that's still yours to do. You can dramatically shift how fast you move between value gates in your work. Saying that the double diamond is dead (not what she said) shows you never understood design in the 1st place.

8

2 Comments

Quite a few designers recently came to me for whiteboard challenge advice. I realized I didn’t have a go-to resource to share — so I wrote one. Whiteboard challenges are one of the most dreaded parts of the interview process. Sitting in front of a panel, being asked to solve a design problem on the spot, while people watch your every move — it sounds stressful. But here’s the truth: whiteboard challenges don’t have to be scary. In fact, they’re often a fairer and more realistic alternative to take-home tasks. They test skills that actually matter on the job: thinking on your feet, structuring a problem, communicating clearly, and collaborating. This week’s newsletter breaks down: - What whiteboard challenges are (and aren’t) - The typical structure you can expect - How to prepare and practice effectively - How to use AI smartly without losing control If you’ve got one coming up — or just want to be ready when the time comes — this is one you won’t want to miss. Make sure to subscribe to not miss this one: https://lnkd.in/eWPDw_wj

47

10 Comments

Why should designers care about the SAAS-pocalypse fear that wiped out $285 billion of market cap last week? For me, investing time to examine signals embedded within headlines help to shed light on uncomfortable truths necesary in chartering new directions, especially during times of change. Last month, Anthropic released "Claude co-work", a series of plug-ins that brings AI intelligence and automation to Microsoft's ubiquitous enterprise software suite. Yes, the same Excel, Power point, etc. which represent the corner stone from which the world’s corporate enterprise businesses are built upon. Claude co-work demonstrated something none of the AI-hype that came before it had done before, it created business justifiable value beyond what most IT sees as "nice to have but non-essential" use cases like generating images for presentations, taking meeting notes and rewriting emails. For the first time, there is a concrete argument for enterprise AI-investment beyond refreshing AI enabled hardware and isolated MS Copilot licenses for "specialty departments" and privileged individuals. The SAAS-pocalypse fear raised a question of "when" and not "if" future agentic AI solutions will eliminate the need for custom software and the sustainability of "per-seat" business model. Correct, the very business model where billions of dollars in enterprise SAAS revenue are generated today… hence the market reaction. UX designers: Will we see enterprise custom software and human "UI-first" design development give way to AI "Agentic-first" solutions being developed? What new challenges await users who manages them? Trend forecasters & human factors experts: Many of today’s roles and associated pain points will cease to exist. How will enterprise & business roles/workflows change and what new pain points or trends do you anticipate shaping the workplace? Industrial designers: Is this finally the catalyst that justifies the timing for an agentic-first hardware product to exist? As a design leader and investor, I've always felt time spent on connecting the dots are essential, they often reveal signals for seismic shifts. so, ask yourself. Are you investing enough time connecting the dots on changes taking place and taking action? Are you identifying new design opportunities for value creation in the face of these changes?

16

Good design needs trust. → Trust to explore. → Trust to test. → Trust to make decisions. When designers are treated like pixel-pushers instead of problem-solvers, creativity shuts down and so does good design.

800

14 Comments