Issue #3: Retail Upskills, Hybrid Biases, and the Prompt Engineering Debate

A few stories on the future of work that caught my eye this week:

1. Why retailers are leading the way on skills. Great feature from the LinkedIn Talent Blog on how large retailers like Walmart and Best Buy are taking skills-first approaches and upskilling their retail talent. The article cites some motivations that make these retailers relatively more inclined to embrace a skills-oriented approach and invest in upskilling (expecting major change due to AI/tech, retail talent attraction and retention, leveraging a large base of talent for potential corporate roles); it didn't cite some others (e.g., keeping unions and unionization at bay, the existential risk when your large retail workforce represent a meaningful amount of your customer base). Regardless, while retail might rank high in its predisposition, programs and policies they've introduced to orient toward skills are notable and instructive for other companies across industries. 📖 Article link
2. Combating bias to make hybrid work work. What do people value the most about hybrid working? "Flexibility." What's the one thing most likely to make hybrid working an exercise in strife and unproductivity? "Distrust"—managers translating working from home as hardly working. It's a fundamental challenge that any company who's adopted a hybrid stance (many companies in most sectors) has been struggling with. Professor Chia-Jung Tsay 's research has put the spotlight on one important dimension companies should recognize: gender bias feeding assumptions on whether people are really working and why they might be away from their desk. In male-dominated or gender-balanced industries (ie most of them), managers are more likely to assume female workers are spending work time on non-work activities (e.g., family, personal) when working virtually or away from their desks in the office. One effective way to combat this, per Tsay's study: share more detailed information to explain absences and remove ambiguity... although that's a tactic vs. getting at the heart of the matter—shifting your culture and approach to support flex time, instead of interrogating it. (h/t Brian Elliott for highlighting this research here) 📖 Article link
3. Settling a common gen AI debate: "prompt engineer - job or capability?" The Hard Fork podcast team interviewed Riley Goodside , who has been working in the machine learning space and doing prompt engineering since before ChatGPT was released broadly. His answer: prompt engineering will primarily a capability, with a small number of jobs akin to solution engineers and oriented toward applying LLMs to new areas (industries, functions). LinkedIn data supports that, with around 800 prompt engineers in the world (250 in the US, 170 in India), concentrated at AI companies like Nvidia, Open AI, and Scale AI (where Riley works). Kevin Roose saw prompt engineering as akin to the digital librarian of the late 90's, helping people figure out how to use the World Wide Web. So as you're working on talent priorities for 2024+, maybe focus less on opening up a new prompt engineering job category and more on developing the skill across your employee base... 🎧 Spotify link