# Tata Elxsi builds real-time industrial safety platform IRIS on AWS

> Tata Elxsi says IRIS, its AWS-based industrial vision platform, cuts unsafe-condition detection from 15–45 minutes to near real time.

- **Topic**: Industry
- **Published**: 2026-09-24T23:25:05.823Z
- **Canonical URL**: https://highsignal.sh/stories/tata-elxsi-builds-real-time-industrial-safety-platform-iris-on-aws-1bceeb4f

## Why It Matters

Industrial plants run hundreds to thousands of cameras that mostly record rather than prevent incidents; IRIS is a vendor-described blueprint for edge-filtered computer vision that scales with camera count without streaming raw video to the cloud.

## Key Findings & Analysis

### What happened

Tata Elxsi, a design and technology services provider spanning automotive, manufacturing, healthcare and transportation, described IRIS (Industrial Real-Time Intelligence System), its industrial vision platform built on AWS, in an AWS Machine Learning Blog post. The post frames the problem as a shift from CCTV as recording to CCTV as prevention: safety teams cannot watch hundreds of feeds at once, so in the customer challenge Tata Elxsi cites detection of unsafe conditions taking 15–45 minutes depending on operator availability, and audit coverage limited to two or three manual walkthroughs per shift. IRIS is built as a serverless, event-driven pipeline that observes at the edge, detects with computer vision, correlates raw detections into safety events, alerts, stores for compliance, and feeds production data back into the system. Video is analyzed at the edge with only safety-relevant frames and structured metadata moving to the cloud; IRIS runs in the Asia Pacific (Mumbai) AWS Region, which Tata Elxsi says was chosen for data-residency requirements and low-latency proximity to customer facilities in India. The edge tier uses AWS IoT Greengrass on GPU-equipped industrial servers such as NVIDIA Jetson AGX Orin or equivalent, connected to the camera network over RTSP/ONVIF. All performance and outcome figures above are Tata Elxsi's and AWS's claims rather than independently verified results.

## Primary Sources & Citations

- [How Tata Elxsi detects industrial safety risks in seconds on AWS](https://aws.amazon.com/blogs/machine-learning/how-tata-elxsi-detects-industrial-safety-risks-in-seconds-on-aws/) — *AWS Machine Learning Blog* (Primary source)

---

[← Back to front page](https://highsignal.sh/) | [Daily Brief](https://highsignal.sh/brief) | [All stories](https://highsignal.sh/latest)