SaeroSense새로센스
  1. Home
  2. Industries
  3. Telecom & IoT

Industry guide

AI for telecom, networking & IoT device fleets.

Turn device logs and telemetry into diagnoses your support team can act on, and ship AI features that run on the hardware you already sell.

The problem

What we hear from teams like yours.

  • Vague tickets, buried evidence

    Customers say “the internet is slow.” The real cause is somewhere in thousands of log lines: WiFi events, DHCP, ARP, reboots, signal strength.

  • Expertise that doesn’t scale

    Diagnosis depends on a few senior engineers who know which patterns matter. Their past conclusions sit in tickets nobody can search.

  • Fleets too big to watch

    Thousands to millions of devices across models and firmware versions. Outages and drift show up as noise until customers call.

What we build

Systems that ship, not demos.

Every build is scoped against a metric you can rerun yourself, and you own the code and models.

  • Log-based fault diagnosis

    Raw device logs are parsed into sessions and causal chains, and come out as ranked, human-readable root causes with the evidence lines behind them.

  • Support copilots

    RAG over device logs, manuals and past incidents, so first-line support resolves issues without escalating, and every answer shows its evidence.

  • Fleet anomaly detection

    Baselines per device model and firmware, with drift and outage detection across the whole fleet, plus alerts your operations team trusts.

  • On-device AI

    Compact models that run offline on embedded Linux. Our WiFi-sensing model is about 230 KB and runs on an ARM device with 1 GB of RAM.

Relevant work

Proof, labeled honestly.

Our peer-reviewed work on power-line fault detection is the same problem as fleet anomaly detection: rare fault signatures hidden in noisy signals, and models that must keep working on equipment they were never trained on.

WiFi log intelligence with RAG: project visualIndustry project

Telecom · home networking

WiFi log intelligence with RAG

Vague “my internet is slow” tickets become a ranked root cause, the log lines that prove it, and the next action.

Result: 7 analysis passes · 3 vector memories

View case study
Industry R&D
93.5%Test accuracy · 230 KB on-device model

Smart home · telecom

On-device WiFi sensing

Routers detect human activity without cameras, fully on the device. A privacy-friendly feature with no cloud cost.

Result: 93.5% test accuracy · 7 activities

View case study
Peer-reviewed · IEEE Access
90.5%Accuracy on unseen power lines

Power systems · transmission

Pattern-recognition auto-reclosing with Bi-LSTM

A fault-detection model that keeps working on power lines it was never trained on. Published in IEEE Access.

Result: 90.5% on unseen lines (GRU: 80.7%)

View case study

Engagement options

Start small. Fixed prices per phase.

Scope first, build only if the numbers say it’s worth it, and add Operate if you’d rather we run it.

Step 1

Scope

Fixed price · 1 to 2 weeks

We study your data and workflow, agree on the metric that defines success, and give you a fixed quote for the build.

  • Data review and an honest feasibility read
  • A success metric you can measure yourself
  • A build plan with a fixed price and timeline
Start with a scope
Step 2

Build

Fixed price · prototype in 2 to 6 weeks, then production

A working prototype on your real data, then hardening, deployment and handover to your team.

  • Working prototype on your data, not slides
  • Evaluation harness you can rerun after every change
  • Deployment to your cloud or on-premise hardware
  • Documentation and training. You own the code and models
Discuss a build
NDA before any data is shared. You own everything.How we work

Common questions

Before you ask.

Q01Our logs are messy and undocumented. Is that a blocker?

No, it’s the normal starting point. Parsing and normalizing mixed log formats is treated as real engineering work in the engagement, not an assumption we skip.

Q02Do we have to send device data to a cloud LLM?

No. Diagnosis pipelines can run in your VPC or on-premise with open-weight models, designed for carrier-grade privacy expectations.

Q03Can it work in real time?

Yes, within honest limits. We usually split the system into a fast streaming layer for detection and a slower reasoning layer for diagnosis, so alerts are timely and explanations are thorough.

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 1 business day with an honest read on feasibility. No slide deck needed.

Send us a message contact@saerosense.com Seoul based · available throughout US business hours