This in vitro study evaluated the accuracy of guided endodontics for the removal of intraradicular fiberglass posts from posterior mandibular teeth and the influence of the operator’s experience in this procedure. Forty root-filled mandibular first molars with an intraradicular fiberglass post were mounted onto 20 mandibular models. Guides for access were made following surface scanning and cone-beam computed tomography (CBCT) using CoDiagnostix software. The models were randomly divided between two operators (n = 20). One was an inexperienced dental surgeon (IN), and the other was a dentist experienced in the guided technique (EX). A postoperative CBCT scan was superimposed on the initial planning, and the means were calculated for the angle and for 3D deviation. The 95% confidence interval (CI) was calculated, and differences between groups were assessed by a t-test. The mean deviation angle was 2.54° (0–5.85°) for IN and 1.55° (0–2.85°) for EX. The mean 3D deviation was 0.44 mm (0.14–0.73 mm) for IN and 0.33 mm (0.11–0.76 mm) for EX. The means of the angle and 3D deviation were significantly different (p = 0.008 and p = 0.049). Guided endodontics was influenced by the operator and allowed access for removing intraradicular fiberglass posts with minimal deviation and greater accuracy for an experienced operator compared with an inexperienced one.