Perception
Multi-sensor fusion across RGB, depth, hyperspectral and X-ray. Segmentation, detection and calibrated uncertainty at line speed.
Careers
Physical AI, where it actually matters.
Most AI work moves pixels. This work moves blades, changes what a plant earns, and takes people off the most dangerous job in food production.
You will ship to a real factory floor, and you will see the number move.
Feedback here is not a dashboard metric a quarter later. It is a yield percentage on a line, measured against the plant’s own systems, the week after you ship.
Teams
Where you could work.
Robotics & control
Cut-path planning, force control, seam-following and grasp planning for deformable product in washdown cells.
Simulation
Omniverse-based carcass and cell simulation, synthetic rare-event generation and CI validation gating.
Plant-edge infrastructure
Rust and C++ runtimes, K3s at the edge, Triton serving, deterministic latency and signed OTA delivery.
Food-safety engineering
HACCP, USDA-FSIS and EU alignment, evidence schemas, validation protocols and audit tooling.
Go-to-market
Solutions engineering, plant commissioning, partnerships with OEMs and integrators, and enterprise sales.
Open roles
Roles open now.
All roles are remote-friendly with regular time on customer plant floors. [PLACEHOLDER]
How we operate
What working here is like.
Small teams, real ownership
You own a workflow end to end — from the sensor mount in the cell to the model, the gate and the evidence record. Nobody hands you a ticket detached from the outcome.
Remote, plant-connected
Work from anywhere, but expect to spend real time in cold rooms with the people who run them.
Equity for everyone
Meaningful ownership at every level. [PLACEHOLDER]
Learning budget
Conferences, courses and hardware you need to do the work well.
Sustainable pace
Commissioning weeks are intense. The rest of the year is not. We plan for both instead of pretending.
Hiring
Our interview process.
-
1
Intro call
Thirty minutes on your background and what you want to build. No brainteasers, no trick questions.
-
2
Technical deep dive
A conversation about real problems from our roadmap — perception, control, simulation or infrastructure.
-
3
Practical exercise
A scoped, paid take-home or a live pairing session. Your choice, and we respect your time either way.
-
4
Team & offer
Meet the people you would work with, ask us anything, and get a decision within a week.
The stakes
Why the work matters.
$25B
Market being rebuilt around autonomy
4.1%
Yield uplift measured on pilot lines [PLACEHOLDER]
61%
Giveaway reduction on fixed-weight packs
0
Foreign-material escapes on monitored lines [PLACEHOLDER]
From customers
The people you would build for.
“The line does not care that every bird is different — Slicium does. We stopped programming a machine and started supervising an operator that adapts to each carcass.”
“Giveaway was the quiet leak nobody could close. Watching the portioner hold target within a couple of grams, shift after shift, changed the economics of the whole pack line.”
“What sold my team was the audit trail. Every reject links back to the frame, the model version and who approved the autonomy level. That is what a recall investigation actually needs.”
Design-partner quotes are composite and pending publication approval. [PLACEHOLDER]
Benefits
The practical stuff.
Remote-first
Work from where you are productive, with travel to plants planned well in advance.
Flexible hours
Commissioning has shifts; the rest of the job does not. Async by default.
Real hardware
Access to edge GPUs, sensors and a test cell — not just a simulator.
Our standards
Engineering with consequences.
SOC 2 Type II (in progress)HACCP-alignedUSDA-FSISEU 853/2004GDPRISO 27001 (planned)
Fail-safe by design
Blade, robot and line stops are deterministic and independent of the cloud. Graceful degradation returns the cell to a safe state if perception confidence drops.
Data residency & on-prem
Run fully on-premise on the plant edge, in your VPC, or hybrid. Sensitive producers can keep every frame inside the facility.
SSO, SAML and RBAC
Enterprise identity, role-based approval gates, and separation of duties between operations, quality and engineering.
Careers FAQ
Questions candidates ask.
No. Most of our team came from robotics, perception, simulation or infrastructure. Domain knowledge is teachable — curiosity about a physical process, and willingness to stand in a cold room at 4am, is not.
It varies by role. Perception and robotics engineers typically spend one week in six on a plant floor during active commissioning; platform roles less. We plan travel well in advance and never spring it on people. [PLACEHOLDER]
Slicium is an independent, founder-led company building original plant-edge autonomy technology. Funding stage, incorporation details and headcount are not published here. [PLACEHOLDER]
Scoped to roughly four hours, paid, and drawn from a real problem we have solved — for example calibrating uncertainty on a detection model, or designing a fail-safe degradation path. You can pair with us live instead if you prefer.
Do not see your role?
If you can help take the blade out of human hands, we want to hear from you regardless of what is posted.