The collective mechanical behavior of cellular aggregates strongly influences morphogenesis, cancer growth and wound healing. While single cell mechanics has been extensively studied, the collective dynamics of a large number of cells inside a tissue is not well understood. Recent in vitro experiments have shown cells in tissues behave like fluids on long timescales and solids on shorter timescales, and exhibit non-gaussian “caging behavior” at intermediate timescales as cells become more tightly packed. These observations are reminiscent of dynamic slowdown and dynamical heterogeneities due to mutual confinement and crowding of particles in disordered, glassy systems. A common and crucial feature of systems with glassy behavior is the existence of a Potential Energy Landscape (PEL) for local rearrangements. For ordinary thermal glassy materials, when these barriers are large compared to thermal fluctuations, its rheology is dependent on the PEL and external mechanical driving. On the other hand, cells in a tissue are non-thermal and overcome energy barriers in the PEL mainly through local active processes, i.e. making new adhesions and cell shape changes. We numerically map the PEL of a confluent tissue as functions of different transition pathways and single cell properties. Analytical calculations are also performed to find the minimal energy shapes for 2-D confluent cell packings.