- Home
- About
About
A small studio that ships.
Saerosense is an applied-AI studio based in Seoul, South Korea. We take AI projects from idea to production system, and the engineers who scope your project are the ones who build it.
What we believe
Most AI projects fail between the demo and the deployment.
Our whole practice is built around that gap: every engagement is scoped against a measurable outcome, every prototype runs on your real data, and every delivery includes the evaluation harness that proves it works — so you can keep proving it after we leave.
We work in English and Korean, on-site in Korea or remote worldwide.
The name
새로센스 — a new sense.
새로 (saero) is Korean for “anew.” Sense is what our systems produce: perception drawn from data that was already there.
Saerosense builds systems that give organizations a new sense — the ability to perceive what their own data has been saying all along: the failing device in a million log lines, the defect in ten thousand images, the answer buried in years of documents.
Team
The engineers in the kickoff are the engineers in the code.
Dr. Muhammad Saad
Engineer with a PhD (South Korea), working across Applied Machine Learning: multi-agent LLM systems, retrieval-augmented generation, computer-vision segmentation, and AI fault diagnosis for telecom-grade networking environments. He has led AI projects from research through carrier-facing production, reporting at CTO level, and publishes research in applied computer vision.
Where this team has shipped
Saerosense is new; the engineering behind it is not.
Our founding team has built systems or published peer-reviewed research in each of these domains.
-
Log fault diagnosis · on-device sensing
Telecom & WiFi
Carrier-grade log fault diagnosis and WiFi-sensing models that run on-device.
-
Segmentation
Infrastructure inspection
Concrete defect segmentation — cracks, spalling, rebar exposure — for scanner platforms.
-
Visual inspection
Manufacturing & safety
Visual inspection, PPE compliance detection, and product authentication models.
-
Segmentation · generative
Medical imaging research
MRI tumor segmentation and generative X-ray enhancement with hospital partners.
-
Fault detection
Power systems
AI fault detection for transmission lines, published in IEEE Transactions.
-
Agents · RAG eval · edge DL
Applied AI research
Multi-agent analysis systems, RAG evaluation, and edge-deployable deep learning.
At a glance
The facts.
We deliberately do not list client logos we cannot verify or projects we cannot discuss; if you want references relevant to your industry, ask us directly.
Start a project
Tell us about your data.
Two or three sentences about the data you have and the outcome you want is enough. We reply within two business days with an honest read on feasibility — no slide deck needed.