Blog

Erik Kokalj

Developer Experience @ Roboflow

Latest Posts by Erik Kokalj

Understanding Video with GPT-6 Astra

11 Sep 2026 • 5 min read

Understanding Video with GPT-6 Astra

GPT-6 Astra has no video input, but a sequence of timestamped frames in one request gets you a timeline of events. Three demos: sports analytics, process monitoring and truck turnaround time.

Unified Namespace (UNS): Add Vision AI Events via MQTT with Roboflow

27 Jul 2026 • 9 min read

What Is a Unified Namespace? Where Vision AI Fits in the UNS

A unified namespace gives every plant system one data model. Roboflow Workflows publish camera detections as MQTT events any MES can read.

Run RF-DETR in NVIDIA DeepStream on Jetson with Roboflow Vision A

22 Jul 2026 • 6 min read

Run RF-DETR in NVIDIA DeepStream on Jetson

From pretrained weights to live multi-camera inference on a Jetson Orin NX: TensorRT engine build, the custom bbox parser DeepStream needs, and per-class colors with a pyds probe.

Gemini 3.6 Flash for Vision on Roboflow Vision AI

22 Jul 2026 • 4 min read

Gemini 3.6 Flash for Vision: Evaluation and Benchmarks

Cheaper and faster than Gemini 3.5 Flash, and the best model we tested on video. Object detection is the one real step back.

How to Scale a Computer Vision Pilot to Production in Manufacturing

15 Jul 2026 • 4 min read

How to Scale a Computer Vision Pilot to Production in Manufacturing

See how to scale a computer vision pilot to production: why 77% stall on integration, not the model, and how to get past it in manufacturing.

Making Visual AI Standard Practice in Complex Manufacturing

8 Jul 2026 • 4 min read

Making Visual AI Standard Practice in Complex Manufacturing

Learn how to implement visual AI in manufacturing from Florida Crystals' quality lead: trust tests, feedback loops, and pilots that scale.

Claude Sonnet 5 vision evaluation on Roboflow

3 Jul 2026 • 2 min read

Claude Sonnet 5 for Vision: Evaluation and Benchmarks

We ran Claude Sonnet 5 through the Roboflow Vision Evals across 67 real vision prompts. It ties Sonnet 4.6, trails Fable 5, and sits well behind Gemini 3.5 Flash, which still leads the board.

Turning Computer Vision Into Real‑World Value at Enterprise Scale

2 Jul 2026 • 5 min read

Turning Computer Vision Into Real‑World Value at Enterprise Scale

Learn why computer vision pilots fail in the real world. Roboflow CEO Joseph Nelson shares a three-step playbook for deploying vision AI in production.

Cost of Poor Quality (COPQ) - improve with Roboflow Visual AI

1 Jul 2026 • 4 min read

Cost of Poor Quality (COPQ): How to Calculate It and Bring It Down

Cost of poor quality is the money defects burn: scrap, rework, downtime, claims. Calculate yours line by line, then cut it with 100% inspection.

Mechanical Turk Closes: How to Move Labeling to Roboflow

1 Jul 2026 • 2 min read

Mechanical Turk Closes: How to Move Labeling to Roboflow

Mechanical Turk closes Sept 30, 2026 and takes Ground Truth's crowd workforce with it. Here's how to export your labels and move to Roboflow in an afternoon.

Manufacturing Vision Systems on Roboflow

1 Jul 2026 • 7 min read

Manufacturing Vision Systems

Manufacturing vision systems explained: the five components, smart camera vs PC-based, rule-based vs AI, and how to choose one for your line.

Human-in-the-Loop Computer Vision

30 Jun 2026 • 4 min read

Human-in-the-Loop Computer Vision: Closing the Gap Between the Model and the Floor

See how human-in-the-loop computer vision closes the gap between model metrics and the floor. Watch operators improve the next model live.

eQMS in Manufacturing with Roboflow vision AI

18 Jun 2026 • 6 min read

What Is an eQMS in Manufacturing?

Discover what an eQMS is and how integrating computer vision automates manufacturing quality control, from real-time defect detection to CAPA tracking.

REST APIs in Manufacturing

16 Jun 2026 • 7 min read

REST APIs in Manufacturing

Learn how REST APIs connect vision AI, MES, and quality systems on the plant floor. Understand when to use REST versus MQTT and OPC UA for industrial data.

Claude Fable 5 for Vision: Evaluation and Benchmarks

11 Jun 2026 • 2 min read

Claude Fable 5 for Vision: Evaluation and Benchmarks

Claude Fable 5 is a strong reasoning model for visual understand but not a state-of-the-art vision model.

MQTT in Manufacturing with Roboflow's vision AI

9 Jun 2026 • 6 min read

MQTT in Manufacturing with Vision AI

Discover how MQTT connects computer vision models to PLCs, SCADA, and MES systems on the plant floor. Learn to automate defect detection and safety alerts.

Electronic Batch Records in Manufacturing with Roboflow Vision AI

4 Jun 2026 • 7 min read

Electronic Batch Records: Verify Data Entry with Vision A

Electronic batch records still depend on people reading gauges and typing numbers. Vision AI verifies each reading before it enters the record.

The CV Integration Stack: How Roboflow Vision AI Fits into PLC, MES, SCADA, and ERP

3 Jun 2026 • 6 min read

The Computer Vision Manufacturing Integration Stack

Connect computer vision to your manufacturing stack: edge inference plus PLC, SCADA, MES, and ERP integration over EtherNet/IP, MQTT, and OPC UA.

Cosmos 3: Evaluation for Vision Use Cases

3 Jun 2026 • 4 min read

Cosmos 3: Evaluation for Vision Use Cases

NVIDIA Cosmos 3 offers zero-shot performance on fixed-camera footage across airports, warehouses, and production lines.

What Is OPC UA? How Factories Connect Vision AI to PLCs with Roboflow

2 Jun 2026 • 7 min read

What Is OPC UA? How Factories Connect Vision AI to PLCs

OPC UA (IEC 62541) explained: what it is, how it compares to MQTT and Modbus, and how it connects Vision AI inspection results to PLCs and MES.

How to Preprocess Data for RF-DETR Training

1 Jun 2026 • 5 min read

How to Preprocess Data for RF-DETR Training

Learn how to rreprocess data for RF-DETR in Roboflow step by step.

YOLO26-Depth and Monocular Depth Estimation

28 May 2026 • 6 min read

What Is YOLO26-Depth? Monocular Depth Estimation

YOLO26 Depth brings metric monocular depth estimation to YOLO.

Gemini 3.5 Flash for Vision: Evaluation and Benchmarks

22 May 2026 • 6 min read

Gemini 3.5 Flash for Vision: Evaluation and Benchmarks

SUMMARY Gemini 3.5 Flash, released at Google I/O on May 19, 2026, currently tops the Roboflow Vision Evals leaderboard across 67 real vision prompts covering defect detection, document understanding, object counting, and spatial reasoning. It outperforms Gemini 3.1 Pro on counting and spatial tasks while running roughly

How to Use EasyOCR Python Tutorial Example on Roboflow Vision AI

20 May 2026 • 7 min read

How to Use EasyOCR

Learn how to use OCR to read and extract text from an image.

Stay Connected

Get the Latest in Computer Vision First