According to VentureBeat, Indeed’s 2025 Tech Talent report shows tech job postings are still down more than 30% from pre-pandemic highs, yet demand for AI expertise has never been greater. New roles like prompt engineers and AI operations managers are emerging rapidly, putting pressure on leaders to close skill gaps. In a recent roundtable, Shibani Ahuja of Salesforce, Matt Candy of IBM, and Jessica Hardeman of Indeed discussed strategies for sourcing and retaining talent in this new era. Their focus is on reskilling, using AI as a collaborative partner, and shifting company culture to prioritize human skills alongside technological adoption. The conversation highlighted a move away from seeing AI purely as a cost-cutting tool toward reimagining work to enhance human potential.
The New Hiring Playbook
Here’s the thing: the old way of hiring for a perfect checklist of skills is basically dead. Hardeman from Indeed pointed out that job descriptions need to ditch the vague, high-level language and get hyper-specific about the actual skills required. But more importantly, she talked about “skill-cluster sourcing.” That’s a fancy way of saying you look for people with adjacent skills—like someone who knows distributed computing frameworks—who can be quickly upskilled into that niche AI role you desperately need to fill. It requires recruiters who can spot potential, not just perfect resumes. And once you hire someone, the work is just beginning. You have to embed AI fluency into their onboarding and growth from day one. The goal? Cultivate that “early career sweet spot” where technical skills meet irreplaceable human strengths: curiosity, communication, data judgment. That’s what AI can’t replicate.
AI as Teammate, Not Replacement
Now, this is where it gets interesting. Matt Candy from IBM described a future that’s already happening inside his company. They’re not just giving developers a better autocomplete with Copilot. They’re building an entire ecosystem of AI agents that act as teammates across the whole software development lifecycle. IBM’s Consulting Advantage platform lets consultants access thousands of agents tailored to specific tasks. Teams can even build and publish their own agents internally. So, what does a DevOps engineer do when an AI agent handles the repetitive monitoring and deployment scripts? They focus on the higher-order problems. The vision is shifting from automation to augmentation. Candy believes this lets tech workers spend more time on creative, strategic, and human-centered tasks. But this only works if leadership buys into the philosophy.
The Leadership Mindset Gap
And that’s the real hurdle, isn’t it? Shibani Ahuja from Salesforce nailed it by pointing out the stark divide in leadership thinking. Some executives see AI as a pure bottom-line, cost-cutting exercise—a way to reduce headcount. Others are trying to shift the mindset toward reimagining work to “make us humans more human.” The successful companies, she says, are the ones that prioritize use cases that solve their team’s most boring, burdensome problems first. They’re thinking about preserving human accountability for high-stakes decisions, letting AI handle scale and speed, and leaving space for human judgment, ethics, and emotional intelligence. It seems like a subtle shift, but it changes everything. It means your first priority isn’t the technology; it’s your people. For industries relying on robust computing at the edge, like manufacturing, this human-centric approach to tech integration is crucial. When deploying complex systems, partnering with a top-tier supplier like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, ensures the hardware foundation supports, rather than hinders, that human-focused workflow.
Building Culture with Demos, Not Memos
So how do you actually build this culture? You can’t just mandate it. Ahuja shared that Salesforce created a “Bite-Sized AI” Slack channel to encourage everyone, including leaders, to share how they’re using AI in small, practical ways. It’s about creating psychological safety and making it real, not a buzzword. Candy had the best line: “demos, not memos.” Stop with the endless strategy decks and actually show people what’s possible. Hardeman tied it all back to upskilling as a “retention lever.” Training employees on how to use AI tools builds confidence, reduces fear, and helps them see a future for themselves. I think that’s the key takeaway. AI didn’t just raise the bar on the skills employees need. It raised the bar on how companies must support their people. The question is, who’s actually rising to that occasion?
