Akangsha Sunil Bedmutha

Akangsha Sunil Bedmutha is an applied artificial intelligence and machine learning leader whose 14-year career has focused on building real-time and batch inference systems across cloud and edge environments, alongside behavioral analytics for large-scale enterprise applications. Her work is defined by the translation of advanced AI capabilities—perception, prediction, and decision intelligence—into production-grade systems that operate under real-world constraints such as latency, reliability, privacy, and cost. Across vision AI, e-commerce search and discovery, marketing intelligence, customer experience, and mobility, Bedmutha has repeatedly taken on roles that go beyond routine delivery, shaping AI product strategy while driving the technical architecture required to scale.

In her current work at Cisco Meraki, Bedmutha leads the vision and execution of multi-modal camera and video intelligence systems deployed across millions of edge devices worldwide. Her objective has been to deliver capabilities such as natural language search, relevance ranking, people counting, cross-camera tracking, object-motion detection, and automated license plate recognition—while meeting strict constraints on compute, memory, and power. The innovation in this work lies in integrating heterogeneous models into a unified edge–cloud ecosystem and enabling low-latency, privacy-aware on-device inference at global scale. Bedmutha has driven lightweight model architectures to reduce latency, built foundational data-platform capabilities to support edge intelligence, and launched analytics features adopted commercially with measurable customer outcomes. She has also helped shape partner co-development initiatives that accelerate delivery timelines and expand addressable customer segments, while serving as a public evangelist for practical Edge AI.

Previously, as Principal Product Manager at Lily AI, Bedmutha focused on improving enterprise search and discovery through AI systems that better interpret consumer behavior and intent. She introduced multi-modal inference pipelines combining computer vision, deep learning, natural language processing, and large language model (LLM)-based techniques to enhance product findability, attribution, and personalization. Her work delivered new product offerings with strong adoption, expanded a flagship attribution platform, and drove measurable improvements in engagement and revenue outcomes for enterprise retailers. She also reduced human-in-the-loop operations by approximately 60 percent through workflow automation, while leading product strategy decisions spanning pricing, packaging, and product-led growth.

At Adobe, Bedmutha led predictive behavioral products for B2B marketers, spanning real-time behavior prediction, send-time optimization, content recommendations, and intent-capture systems. Her objective was to operationalize predictive audience capabilities that improved marketing performance at scale. These programs were notable for converting large-scale behavioral modeling and NLP-driven information retrieval into enterprise-ready solutions, enabling rapid onboarding of customers within a year of launch and delivering measurable improvements such as higher campaign conversion and lower opt-outs. Her contributions included establishing A/B testing frameworks, defining feature sequencing strategies, and introducing NLP-based similarity and clustering approaches for content discovery and targeting.

Bedmutha’s work in mobility and customer analytics further demonstrates the breadth of her applied AI portfolio. At Rivian, she built analytics capabilities that anticipated buyer behavior across the electric vehicle purchasing lifecycle, identifying cancellation risk and enabling retention interventions. By developing behavioral prediction pipelines, churn anticipation analytics, and demand forecasting foundations, she supported a measurable reduction in order cancellations within a short timeframe. Earlier, at Deloitte, she delivered AI and data strategy initiatives aimed at reducing operational bottlenecks in high-traffic service environments. By combining predictive demand modeling with operational optimization, she launched a resource-allocation solution that reduced wait times and improved customer satisfaction, influencing executive strategy through evidence-based recommendations.

Bedmutha’s early work at Yahoo (via Persistent Systems) focused on search relevance and behavioral analytics, strengthening the search experience through deeper modeling of user intent and algorithmic optimization. Her analyses contributed to measurable improvements in engagement metrics such as click-through rates and informed product direction for leadership stakeholders.

Across these roles, Bedmutha has consistently demonstrated the ability to bridge AI research methods with product execution—delivering systems that operate at scale, improve decision quality, and create durable customer value. Her record reflects sustained professional excellence in inference systems, behavioral analytics, and multi-modal AI applications, supporting strong alignment with the expectations of Fellowship membership in the International Institute of Computer Science Professional Association (IICSPA).