Laborup

Here is a fact that is both obvious and, once you sit with it, slightly absurd. A 24-year-old who spends her days in a spreadsheet at a bank has a LinkedIn, a headshot, an endorsement or two, and a search-engine footprint. A 24-year-old who spends her days running a five-axis CNC machine that makes a part for a jet engine has, professionally speaking, almost nothing. She is invisible to the internet. When a factory 40 miles away needs exactly her, there is no good way to find her, and no good way for her to be found.

Laborup, a startup out of Knoxville, Tennessee, exists to fix that particular market failure, and the way it describes the problem is worth quoting because it is the entire pitch in two sentences. "LinkedIn has helped white-collar workers establish digital identities," the founder likes to say. "However, blue-collar workers remain virtually invisible online." If you believe that sentence is true - and it is - then there is a very large, very valuable database that simply does not exist yet, and someone is going to build it.

The someone is Tasimba Jonga, who goes by Simba, and his origin story is the kind that venture capitalists put in the first slide. He grew up between Zimbabwe and Tennessee, studied chemical engineering at the University of Tennessee, did stints doing manufacturing engineering at Bayer and Dow, and then got into Stanford's Knight-Hennessy program to study artificial intelligence. And then he did the thing you are not supposed to do: he left. Left Stanford, left the Bay Area, and moved back to Knoxville, on the theory that if you want to fix how factories hire, you should probably live near some factories.

That line - talent as a supply chain you route, rather than a résumé pile you sift - is the intellectual core of the company. Manufacturers have spent 40 years optimizing the movement of physical things: just-in-time inventory, logistics software, the whole apparatus. They have spent approximately zero of that energy on the movement of people, which is strange, because the people are the constraint. A factory can have the machines and the orders and still be unable to run a second shift because it cannot find four more maintenance techs. Laborup's bet is that the labor side of the factory deserves the same software the parts side got.

What it actually does

So what is the product? At the worker end, it is disarmingly simple. Instead of asking a machinist to fill out a form - and machinists, reasonably, do not want to fill out forms - Laborup points a voice-first AI agent at them. The agent has a conversation. It asks what you can do, and because it has an index of more than a million skills, it knows the follow-up questions: what materials, what tolerances, what machines. About five minutes later, out comes a structured profile and a resume that did not exist before. The worker did not write it. They just talked.

At the employer end, it is a search problem plus a guarantee. Manufacturers get access to a vetted, pre-screened network of skilled workers, and rather than pay to post a job into the void, they pay for outcomes - successful hires. Laborup layers human recruiters on top of the AI, calls the combination "agentic recruiting," and wraps it in a backfill guarantee: if the hire doesn't stick, they replace them. The company claims hiring that is 5 to 10 times faster, cost-per-hire down around 70 percent, and retention two to three times better. Those are the company's own figures, from its own customers, so read them the way you read any vendor's numbers - directionally, with an eyebrow slightly raised.

But the mechanism underneath is the interesting part, and it is the part that makes this an AI company rather than a job board. Laborup says its matching learns from real outcomes: who showed up, who was still there at 90 days, who got promoted. A job board optimizes for clicks. Laborup is trying to optimize for the thing employers actually care about, which is whether the welder is still welding six months later. That is a much harder signal to collect, and if the company genuinely collects it, that data is the moat.

The money, and who is behind it

The financing tells you who believes the thesis. Laborup raised a roughly $1.9 million pre-seed in 2024, then closed a $5.8 million seed round in 2025 led by Norwest Venture Partners, bringing total funding to about $7.7 million. The round list is where it gets fun. Alongside the institutional names - Torch Capital, Threshold Ventures, Heartland Ventures - the angel roster includes Jeff Dean, who is the chief scientist of Google's AI effort, and Evan Moore, a co-founder of DoorDash. An earlier check came from Jeff Jordan of Andreessen Horowitz.

Now, angels invest in lots of things, and you should not over-read a single check. But there is a pattern here worth naming. DoorDash is, underneath the app, a marketplace that routes labor to where it is demanded in real time. Google's AI leadership presumably has views on whether a voice agent can usefully interview a tradesperson. The people writing these checks are, in their day jobs, experts in exactly the two things Laborup is trying to fuse: labor marketplaces and applied AI. That is either a coincidence or a signal, and it is probably a signal.

Why Tennessee, and why now

The geography is not an accident, and this is the part where the macro story does real work. American reshoring - the long, halting effort to build things domestically again - is happening most visibly across the South and the Midwest. Those regions are where the new plants are going, and they are where the skilled-labor shortage bites first and hardest. Laborup planted itself in East Tennessee, grew a vetted network from around 10,000 workers to more than 25,000, and has since pushed into Chattanooga with a stated ambition of six figures of workers across the state. The strategy is hyper-local density before geographic breadth, which is the correct way to build a marketplace: get liquidity in one metro before you spread thin across ten.

There is a national-security framing the company reaches for too - the idea that a country that cannot staff its factories cannot defend itself - and whether or not you find that overwrought, it is doing something clever rhetorically. It reframes a staffing startup as infrastructure. And its early customers lean that way: aerospace and defense shops, an outfit that makes precision tooling, a case-and-container manufacturer. These are not gig-economy dabblers. They are companies for whom a single unfilled maintenance role can idle an expensive line.

The honest caveats

A few things are worth flagging, because a good profile is not a press release. Laborup is early - about 19 people, a couple of years old, its impressive metrics drawn from a handful of named customers in one region. The distinction it draws from the gig-labor incumbents like Instawork and Veryable is real but requires proving: staffing skilled permanent trades is a genuinely different, and harder, business than filling a warehouse shift. And the outcome-based data flywheel that would make the whole thing defensible is exactly the kind of thing that is easy to describe and hard to accumulate. None of that is a knock. It is just the difference between a good story and a proven one, and right now Laborup has clearly got the first and is working on the second.

Still, the underlying observation is hard to argue with. The internet built professional identity for the people who work at desks and skipped everyone else. There are 10 million skilled workers on the other side of that gap, holding the actual tools, and someone is going to build the layer that finds them, verifies them, and moves them to where the work is. Laborup has decided to be that someone, from Knoxville, with a voice agent and a gear for a logo. It is, at minimum, a very specific bet on a very real problem - and those are usually the ones worth watching.