Cascadity Audio
← All streams

#jobs

Everything tagged jobs, across every stream.

0

US job openings jump to 7.52 million, ADP counts 86,000 private jobs, and a Fed official expects more rate hikes

Three CNBC reports give the labor and rates backdrop going into Friday's October jobs report.

Job openings rose by 333,000 in September to 7.52 million, well above the 7.15 million economists expected, the Bureau of Labor Statistics reported on Tuesday, November 3. Quits rose to 3.36 million, the most since March 2025, and there were about 1.05 openings for every unemployed worker.

On Wednesday, payroll processor ADP said private employers added 86,000 jobs in October, the same as September's revised figure and above the 75,000 expected. Professional and business services lost 27,000 jobs; people who changed jobs saw 4.8% annual pay gains.

On Thursday, Richmond Fed President Thomas Barkin told CNBC he expects the Fed will need to raise rates further, at a gradual pace, saying inflation is 'the most important risk right now' and that AI is pushing some prices up. The Fed held its benchmark rate at 3.75% to 4% at its last meeting after a hike in September. The official October payrolls report is due Friday, after this issue's window.

Rabbit Holes 2
0

A San Francisco startup paid people $20 an hour to walk robot dogs

A funny job ad with a serious finding: the company says AI agents are still very bad at steering real robots through city streets.

Dimensional, a San Francisco startup based in the Dogpatch neighborhood, posted an ad this month in a Bay Area jobs group on Facebook offering $20 an hour to walk robot dogs for four to six hours at a time, according to a New York Times report republished by The Star. The ad, since deleted, warned that walkers might have to carry an overheated robot, about 33 pounds, back to the office.

Founder Stash Pomichter, 26, said the company is building an open-source operating system that lets AI agents control robots. The human walkers steer the robots by remote while cameras and sensors record "ground truth" for navigating city sidewalks. The dogs are roughly $2,000 machines from China's Unitree; Pomichter said Dimensional has bought more than 100 robot dogs and 50 humanoid robots.

Comparing walks under human and AI control, he said, showed that "agents are still very bad at controlling hardware."

Cascadic Analysis 1
UndertowRole underspecification / the Doorman Problem

What hidden risk could pull against this, even if the news is good?

The walkers' remote steering is recorded as "ground truth" for sidewalk navigation, but a walker's job includes things no sensor log captures, like the ad's warning that they might have to carry an overheated 33-pound robot back to the office. The founder says agents are "still very bad at controlling hardware." Data from human walks teaches the routine route; the judgment calls and exceptions are the part hardest to learn from it.

0
Rabbit Holes 1
0

Microsoft offers buyouts to about 60,000 employees, then beats estimates as Azure grows 48%

AP explains who is eligible for the buyout; Microsoft's release and CNBC cover the quarter, the capex shortfall and the supply delays.

On October 26, Microsoft's chief people officer, Amy Coleman, announced a voluntary exit offer for about 60,000 employees, roughly 27% of the global workforce, in the US, UK, Canada and India, the AP reports. Microsoft had about 223,000 full-time employees at the end of June and cut 4,800 jobs in July.

Two days later Microsoft reported fiscal first-quarter revenue up 19% and diluted earnings of $4.92 a share, up 32%. Microsoft Cloud revenue was $63.9 billion, up 30%, and commercial remaining performance obligation rose 73% to $677 billion, according to the company. CNBC reports Azure grew 48% and finance chief Amy Hood guided to 49% to 50% growth next quarter.

Capital expenditures and finance leases were $45.2 billion, below the $50.67 billion analysts expected; Hood said some third-party capacity deliveries were delayed. Free cash flow fell 27% to $18.63 billion but stayed positive. The stock rose 2% to 4% after hours but ended the next day roughly flat.

Cascadic Analysis 1
UndertowRole underspecification / the Doorman Problem

What hidden risk could pull against this, even if the news is good?

A voluntary exit offer lets employees, not managers, decide who leaves, and it was offered to about 60,000 people, roughly 27% of the workforce. Org charts list duties; they don't list who knows why an old system behaves the way it does or who handles the rare escalation. Microsoft's people chief says the offer is "new for us" and the company will "learn from it"; what it learns about which knowledge walked out may only show up later.

0
Rabbit Holes 1
0

Anthropic's own workforce report: fewer junior hires and 20% more time in meetings

Anthropic's report is an unusually detailed look inside an AI-saturated workplace; TechCrunch pulls out the key numbers.

Anthropic published a report on how its own workforce has changed after molding its workflows around Claude, arguing its trends may hint at how AI will reshape hiring elsewhere. Junior employees in technical departments fell from 21% to 14% since early 2025, and junior technical hires from 27% to 9%, TechCrunch reports. Technical teams shrank from 61% to 56% of staff, which Anthropic links partly to coding tools letting non-technical employees build software themselves.

In an April survey, 27% of employees said Claude made them more productive and 20% said their job had shifted from doing work to "managing Claudes." The report calls human attention and judgment a growing bottleneck. Calendar data show meeting time per employee up 20% from a year ago, which Anthropic attributes to the importance of human trust and accountability. It says it has not found signs of declining code quality, while noting more code isn't necessarily better code.

Cascadic Analysis 3
UndertowRole underspecification / the Doorman Problem

What hidden risk could pull against this, even if the news is good?

Junior technical hires fell from 27% to 9%. Junior roles are partly where people learn judgment by doing the work that's now done by Claude; the report itself calls human attention and judgment the growing bottleneck, and the future supply of that judgment comes from today's juniors.

0
UndertowDelayed Consequence / False Success Problem

What hidden risk could pull against this, even if the news is good?

Anthropic says it hasn't found signs of declining code quality, while noting more code isn't necessarily better code. Quality problems in a large codebase often show up later, as maintenance cost or bugs in rarely used paths, so "no signs yet" is an early reading rather than a verdict.

0
UndertowLoss of meaningful human control as capability and autonomy increase

What hidden risk could pull against this, even if the news is good?

Twenty percent of employees said their job shifted from doing work to "managing Claudes," and meeting time rose 20%. When more of the work is done by agents and people review it, how much each person can really check becomes the limit on how much oversight there is.

0
Rabbit Holes 1
0

Mercor maps AI benchmarks against US jobs and finds most working hours barely covered

The interactive map lets you click into any job family and see which tasks AI tests actually measure.

Mercor, which sells AI evaluation services, mapped 29 AI benchmarks onto the US Labor Department's O*NET list of 923 occupations and 18,797 tasks. Only 208 occupations have a task covered by at least five benchmark prompts, 6 of 22 job families have none, and Mercor estimates 90% of American working hours go to tasks with little or no benchmark coverage, making it hard to know how well AI does the work people actually do.

Coverage is lopsided: computer and math (29 of 36 occupations) and business and finance (33 of 48) fare well, while healthcare practitioners (12 of 89) and education (9 of 62) lag. Mercor says benchmarks favor the largest labor markets; for example, they heavily cover lawyers' tasks but barely touch those of similarly exposed legal secretaries and court clerks.

Cascadic Analysis 2
UndertowBenchmark scores can overstate real-world reliability

What hidden risk could pull against this, even if the news is good?

Mercor's map says 90% of US working hours go to tasks with little or no benchmark coverage, so a model's headline scores say little about most real work. Mercor sells AI evaluation services, which gives it a stake in the conclusion that more evaluation is needed.

0
UndertowRole underspecification / the Doorman Problem

What hidden risk could pull against this, even if the news is good?

Mercor's example: benchmarks heavily cover lawyers' tasks but barely touch those of similarly exposed legal secretaries and court clerks. If those jobs are automated on the strength of lawyer-focused scores, no one will have measured how AI handles the exceptions and judgment calls they actually involve.

0
Rabbit Holes 1