Machine vision offers an objective, replicable and fast way for inspection and such outperforms manual quality assurance. Deep-learning anomaly-detection is the method of choice in electronics industry. Our software is based on the Halcon vision library offering cutting-edge algorithms of artificial intelligence. The system detects flaws like missing or misplaced components, bended wires, or stains. For large boards, we offer a system with a high-resolution camera that discerns even the smallest defects. The operator easily trains new test specimens and receives a clear monitoring of the produced quality.