Rainfed agro-ecosystems, the purported grey patches untouched by the Green Revolution or most technological advances, occupy a prominent position in Indian agriculture. Cropping intensities and crop yields are low and unstable in these areas due to unpredictable patterns of rainfall, a host of biotic and abiotic stresses and adherence to traditional farm practices. This precarious food security situation is especially dangerous in the central Indian tribal belt (also known as the poverty belt) which is a typical rainfed area dominated by tribal communities. More than 90% of the tribal people are totally dependent on agriculture and produce much of what they eat. Small land holdings and their low productivity, along with uncertainties in rainfall patterns, increases economic and social risks for these farmers. With degraded soils and unreliable weather patterns, return on investment is uncertain and likely to be much lower overall than under irrigated conditions with better soils. Under such conditions, one approach to achieve improved crop production is to minimize soil and other natural resource degradation by adopting a set of crop-nutrient-water-land system management practices, such as conservation agriculture (CA). To assess the effect of introduced technology under local ecological and socio-economic conditions, the study focused on two ecosystem services: a) provisional, and b) regulatory through five treatments consisting of farmers' traditional practice (FP) which was conventional tillage with broadcast of local variety maize (Zea mays L.); and four CA treatments viz., conventional tillage with sole cropped maize using line sowing of the improved maize cultivar 'Nilesh' (CT-M); conventional tillage with maize intercropped with the improved cowpea (Vigna unguiculata L. cultivar 'Hariyalli Bush') (CT-M + C); reduced tillage with sole cropped maize (MT-M); and reduced tillage with maize + cowpea (MT-M + C). After harvest of maize and cowpea, mustard was planted as a post rainy season crop and all the mustard plant residues were returned to their respective plots as residue cover except FP. Under provisional ecosystem services, performance of CA on crop yield, and profitability was assessed through maize equivalent yield and partial budget analysis, respectively. Results showed that reduced tillage combined with maize-cowpea intercropping (MT-M + C) followed by mustard residue retention had higher system productivity and net benefits, an increase of 200% and 230%, respectively over FP. Under regulatory ecosystem services, the soil quality was assessed through calculation of soil quality index (SQI) which was highest under MT-M + C followed by mustard residue retention and lowest under farmers' practices. In terms of CA treatment preference, 35% of the farmers indicated a strong preference for MT-M + C compared to 14% for FP. Combined, these results clearly demonstrate the potential of CA to simultaneously increase yield, diversify crop production and improve soil quality which should support a move towards sustainable intensification of crop production to improve future household income and food security. Additionally, using a transdisciplinary approach fully engaged all stakeholders in co-designing the CA treatments appropriate for the farmers and local environmental conditions leading to significant impacts on economic livelihoods, environmental sustainability and food security.