In endovascular interventions, thin, flexible instruments are inserted through the skin into the blood vessels to diagnose and treat various diseases of the vascular system. One drawback is that the instruments are difficult to maneuver in the desired direction due to limitations in shape and flexibility. Another disadvantage is that the interventions are performed under intermittent fluoroscopy/angiography imaging. Magnetic resonance imaging (MRI) may offer advantages over X-ray guidance. It presents a good soft tissue contrast without the use of nephrotoxic media or ionizing radiation. The aim of this study is to develop a guidewire that is compatible with MRI and includes a steerable segment at the tip. This added degree-of-freedom may improve the maneuverability of the devices thereby the efficiently and safety of the navigation. A 1.6 m (5 ft, 3 in.) long and 0.035 in. diameter guidewire that consists of MR compatible materials and has a flexible tip was designed. The only metallic part was a nitinol rod that was implemented at the distal flexible tip. To limit the risk of heating in the MRI, this rod was kept shorter than 30 mm. The tip could be deflected in one direction by pulling on a Dyneema wire that was placed in the lumen of the shaft of the guidewire. To drive the steerable tip, a handle that could be easily attached/detached from the instrument was designed and implemented. Using the handle, the tip of the 1.60 m long guidewire prototype could be actuated to reach angles from 30 deg to 250 deg. The handle could easily be placed on and removed from the guidewire, so conventional 0.035 in.–compatible catheters could slide over from the proximal end. However, in order to make the guidewire more efficient to enter a bifurcation, the stiffness of the tip should progressively increase from its proximal to its distal end. The guidewire was imaged in a 1.5T MRI using real-time imaging without producing artifacts that would have shaded the anatomy. It was possible to assemble a guidewire with a steerable segment in the required size, using MR compatible materials. Therefore, the current design is a promising proof of concept and allowed us to clearly identify the features that need to be improved in order to come to a clinically applicable instrument.