Extreme-mass-ratio-inspirals (EMRIs) are test beds of strong field gravity and will be targeted by LISA. Their waveforms, with the required degree of accuracy for detection, are not yet known. In collaboration with Kostas Glampedakis and Stanislav Babak, I am pursuing a line of attack that permits fully relativistic description of the orbital motion and the spacetime. This approach relies on the assumption of a ``quasi-Kerr" massive object, which is an object with slightly different multipole moments (for the same mass and spin) than those of a Kerr BH. By utilizing the Hartle-Thorne metric it is possible to find the leading order quadrupole moment deviation. All higher moment deviations are neglected. The advantage of working with a quasi-Kerr spacetime is that it allows the derivation of fully relativistic orbital equations motion and the description of realistic orbits (which are eccentric and inclined with respect to the central object's equatorial plane). It is then possible to construct approximate semi-relativistic ``kludged'' waveforms that are sufficiently accurate for data analysis simulations (just like the Kerr kludged waveforms being used now). As a final step, we aim to obtain a ``quasi-Teukolsky" equation and compute accurate waveforms to replace the above-mentioned kludged waveforms.