The AI Hiring Paradox:
More candidates will be available.
Fewer candidates will be highly desirable.
AI is splitting the market into two groups:
People who have rebuilt how they work around AI.
People still operating the way they did a year ago.
The AI-native group brings systems, workflows, and habits that raise the output of everyone around them. They change how teams build, ship, validate, and operate.
That’s the paradox.
Availability will rise.
Desirability will concentrate.
The market will be highly competitive while many candidates are available but are being sidelined.
For those not watching: LA mayor’s race is the blueprint for every campaign that comes next.
This is the first truly AI-native political race.
Fictional ai generated videos that feel real. Supporter-generated narratives that move faster than campaigns can react. Candidates turned into characters.
From 2010 to 2025, the “social media candidates” won.
In 2026 and beyond, the “AI candidates” will win.
The campaigns that understand this first will define reality before their opponents can respond.
LA is the future. Get ready for the midterms 🍿
There is an inherent conflict in OpenAI and Anthropic building their own deployment teams.
Enterprise AI adoption depends heavily on implementation, so the move makes sense.
But model companies are not neutral advisors.
Their incentive is to drive usage of their own models.
The right answer for most enterprises will be multi-model, workflow-specific, and cost-aware:
Claude where Claude is strongest.
OpenAI where OpenAI is strongest.
Open source where cost/control matter.
No LLM where software or rules are enough.
I highly doubt these companies will consistently prioritize what is best for customers when it comes at the expense of their own model adoption.
@BestBuy should revamp their Geek Squad to providing AI services to mainstreet USA. Sell Hardware + Software Consulting for local SMBs. They have the name, brand awareness, and great people. They just need to be re-trained, repurposed, + new biz model for today. Min 20% lift on stock on announcement.
@btaylor This is great. Lots of changes need to be made to interview process for the new era of AI-native work.
We recently launched the first AI Fluency Standard for global engineering hiring, as a rubric to understand the level so AI Fluency for engineers.
https://t.co/Si1OPxkwX1
@levie A lot of consulting firms are selling the HOW to implement AI.
90%+ of businesses still don't know WHAT AI can do for them.
The gap is still biggest around understanding what is possible, not just how to deploy.
OpenClaw is the fastest-growing open source project, but there are no stories of running it safely in production at scale. As we started deploying agents internally at @brexHQ, we couldn’t stop thinking about this question.
Agents work, but nobody wants to give them real credentials. Instead of waiting for a solution to emerge, we decided to try a novel approach: using LLMs to judge the network traffic of an AI agent.
Today we’re announcing CrabTrap, an open-source proxy that intercepts every outbound request and blocks risky activity using LLMs, before it ever hits an external API. The results are promising; we believe it’s a meaningful step forward in the security of agent harnesses in production environments.
Try it out today.
(As a side note, it was really fun to work personally on a real systems problem again. And btw, if you want to work at a place where the CEO is building proxies at night, we’re hiring!)
🚀 Launch Day 🚀
Software delivery is now in the AI-native era.
Writing code = commoditized.
The new edge is agentic orchestration.
That changes how you interview and who you hire.
And that's why Terminal just launched the first AI Fluency Standard for hiring.
Watch below 👇
@andrewchen For early builds PM led coding can get you surprisingly far. In scaled products the leverage flips. You want engineers owning more product and design not product trying to code around complex systems. 1 arch doesn't solve / becomes bottleneck.
Me: So how’s AI impacting your teams?
CTO 1: It’s helping.
CTO 2: It’s exposing.
CTO 3: It’s like giving a chainsaw to a toddler.
CTO 4: Just pass the wine.
Loved sitting down with @JTLonsdale to talk all about what we're building at Terminal, current state of engineering hiring, and the opportunity of global talent.
thanks for having me on!
How is AI augmenting software engineers? Will it replace or commoditize certain roles?
@JTLonsdale sits down with @dserota, co-founder & CEO of @JoinTerminal, to talk tech hiring trends & the global competition for talent.
Got to sit down for Episode 11 of @Shore_Stories with @dserota, CEO and Co-founder of @JoinTerminal - a platform for hiring engineers remotely.
We covered a lot of ground in recruiting trends, and specifically about LatAm.
Here are some interesting takeaways. 👇
Today I got called out by the legendary @garrytan for some unfortunate bugs he’s been experiencing in Frame. Oof, he was right.
Because of that, I decided it’s time to let you all in on a big secret. If you’ve been observing https://t.co/0PDma013C0 from the outside, you may have noticed our product updates have slowed down and have wondered what’s been going on.
I’ve witnessed this myself with other products I’ve been a customer of. A once innovative company you love seems to quietly fade off. It often means they’ve lost their steam or frankly stopped caring. Is that what’s happening to the Frame product? Hell to the no!
Secretly, internally, we’ve had a roaring engine of innovation and we’re building the BEST stuff in the 8 year history of Frame. If you could walk inside the virtual Frame office, you’d feel the heat immediately because the team has been COOKING.
For the past 3 years we’ve been working on a brand new version of https://t.co/0PDma013C0, truly redesigned from the ground up. It’s the most important work of my professional life and we plan on sharing it with all of you sometime early next year.
We didn’t think it would take this long but alas, here we are.
We’ve set insane goals for ourselves on not only the product features we’re delivering but also around product craft, quality, and reliability. When it comes to truly delightful products, we’re striving to set a new bar for the industry.
I love how products like @linear and @arcinternet are continuing to drive up our expectations of great software and we fully expect the next version of Frame to to live amongst them.
Because we’ve been dedicating SO much time and energy into the new version of our product, I acknowledge it’s led to an increase in bugs of our existing product.
We have a swat team working to reduce some of those pesky bugs and ultimately I’m confident this will all be worth it. Today, we’re smack in the messy middle.
As I write this message, we have hundreds of designers, engineers, and product people who are tirelessly working to bring this mission to life. Huge shoutout to all the Frame team who are reading this.
I also have to say, our colleagues at Adobe have been SO supportive. What an incredible company to be a part of. There’s a lot of exciting work we’re doing with Frame, across Adobe Creative Cloud.
In many ways, I think Frame was a pioneer in high quality creative software on the web and I have to tell you, the new frame is fucking awesome.
To all the old and new Frame supporters out there, we’ve been building this new version for you. Thank you 🤍🤍
The US, unfortunately is far too burdensome for something like this to happen with any speed. But we've (@JoinTerminal) helped dozens of companies move amazing engineers from all over the world into Canada via our platform/local entity, and their global talent stream program. We put a lot of investment into Canada and the local ecosystem to allow for this. Happy to chat!
I'm putting together an upcoming webinar for talent execs from mid-late stage tech startups. Here is my outline so far - what else should I add/remove/adjust?
And like if you want to an invite/copy of recording.
Top Global Hiring Trends: Data and Insights from Terminal
1. Snapshot on macro hiring data
- Where are we in the hiring cycle, what type of hiring is heating up, where are the biggest gaps in labor force (tech related)
2. AI's impact on technical hiring
- Does AI really replace the need for as many developers in headcount plans?
- How do technical assessments change in post GPT world?
- Where can AI have the most impact in the hiring processes and what are the best tools out there?
- Where in the world is the best AI talent?
3. Terminal's insight - Contract-to-hire Takeover
- Global paradigm on contract-to-hire models
- Economic risk adjustment / RIF hangover
4. Status of Remote vs. Hybrid vs. Distributed vs. In-person
- What evidence have we seen on impact of different models in the three years since the start of the pandemic