Kunal Rao is a Researcher in the Integrated Systems Department at NEC Laboratories America in Princeton, New Jersey. He earned his Bachelor of Technology from COEP Technological University and his M.S. in Electrical and Computer Engineering from the University of Florida.
Kunal has led multiple research and development initiatives from concept to deployment, spanning early system design, prototyping, customer-facing proofs-of-concept, and production-ready implementations. His work has resulted in several technical recognitions and awards. His recent work focuses on AI/ML-driven systems, including Generative AI applications and platforms. More broadly, his research interests include distributed systems, edge and cloud computing, real-time processing, video analytics pipelines, graph computing and analytics, and middleware for high-performance, parallel, and heterogeneous computing.
Over the course of his research career, Kunal has authored or co-authored more than 30 peer-reviewed conference and journal publications. He is also an inventor or co-inventor for over 30 granted U.S. and international patents, with many additional patent applications pending.
Generative AI is transforming cloud computing. At Cloud Computing 2026, Kunal Rao will chair the GenAI4Cloud track and deliver a keynote on software engineering in the AI era, exploring how AI agents, LLMs, and intelligent infrastructure are redefining the cloud stack.
Deploying microservices across the computing continuum (edgecloud) requires placement decisions that adapt to workload variation and heterogeneous infrastructure, yet existing solutions often rely on static policies or opaque heuristics. We present Bellona a system for reliable and auditable Large Language Model (LLM)-driven workflow execution that combines a declarative specification language with a runtime that orchestrates tool calls, conditional control flow, and structured LLM reasoning. Using Bellona, we implement an agentic placement workflow that automatically recommends edge or cloud execution. The workflow uses structured prompts and verifiable tool interactions to (i) parse placement and latency-report instructions, (ii) update the latency log, and (iii) select placements based on measured latency improvement thresholds. We evaluate the resulting agent on two representative microservices-based video analytics applications (human-attributes detection and face recognition) over two days of varying workload. Across 1,440 placement decisions per service, the agent achieves accuracies of 94.66%/84.94% (human-attributes detection, Day1/Day2) and 80.91%/96.53% (face recognition, Day1/Day2) with GPT-4o; with GPT-5, accuracy increases to 98.82%/99.45% (human-attributes detection) and 99.31%/99.8% (face recognition). These results demonstrate that Bellona can support practical, self-improving agentic control for placement of microservices on the computing continuum.
Large language models (LLMs) are becoming the centerpiece in the design and deployment of Agentic artificial intelligence (AI) systems. AI agents typically have (a) reasoning ability to analyze and think through the given task, (b) context/memory to remember things in the short-term and long-term, and (c) tools at their disposal to interact with the outsideworld. While solving the given task, it must decide whether tool use is required; if so, it must then select the appropriate tool and invoke it with the correct parameters. Although LLMs have advanced considerably in recent years, their tool-use capabilities remain limited. Even OpenAIs most capable model to date, GPT-5, continues to struggle with reliable tool usage. In this paper, we propose TacTool, which empowers AI agents with improved tool selection and tool call formulation using different LLMs. We conduct experiments using Nestful and Berkeley Function Calling Leaderboard version 3 (BFCLv3) benchmarks and show that TacTool achieves ?27% and ?3% improvement over GPT- 4o on Nestful and BFCL v3 dataset, respectively.
Identifying subject-matter experts within organizations remains a challenging task due to the scale, heterogeneity, and unstructured nature of enterprise knowledge assets. We present TalentScout, an AI-driven expert identification system that constructs a unified, skill-centric knowledge graph by ingesting and analyzing diverse media, including research papers, reports, presentations, transcripts, and supervisor recommendations. TalentScout’s modular architecture integrates document parsing, audio/video transcription, metadata extraction, large language model-based skill extraction, multi-factor author disambiguation, and evidence-weighted skill attribution. At query time, TalentScout decomposes natural language queries into canonical skill requirements, traverses the constructed knowledge graph, and ranks experts based on aggregated skill weights, document quality, and endorsement signals, providing document-level justifications for each recommendation. We evaluate TalentScout on multiple public and internal enterprise datasets, including DBLP, TREC Enterprise, Tilburg, and ManConCorpus. Using standard information retrieval metrics such as Precision@ 5, Recall@5, nDCG@5, and Mean Reciprocal Rank (MRR), TalentScout consistently outperforms leading baselines, achieving up to 24% higher Precision@ 5 in early expert retrieval. The results highlight TalentScouts scalability, transparency, and accuracy, establishing it as a practical solution for evidence-based expert discovery and organizational talent management.
Creating effective slide presentations requires adapting both content and structure to match the communication context e.g. whether the presentation is for summarizing to executives, or reporting progress to research supervisors. In research and enterprise environments, this need for context-sensitive presentations often leads to repeated, manual reformatting of the same material to suit different audiences. Existing generative systems support slide creation but typically rely on structured inputs, assume a fixed format, and offer limited ability to iteratively refine outputs through natural language feedback. Moreover, they rarely accommodate organizational constraints such as formatting guidelines, domain-specific terminology, or branding requirements. We present SlideCraft, a context-aware generative agent that autonomously creates and edits slide presentations based on natural language instructions. SlideCraft infers the intended presentation context, such as an executive-facing or a project review summary for technical oversight, and selects the appropriate slide template. It then synthesizes content from input documents, enriches it with external knowledge and internal assets, assembles it into a structured intermediate representation, and generates a validated slide deck. SlideCraft supports both first-time slide creation and iterative updates, operating through familiar natural language interfaces like email or messaging tools. Our experiments demonstrate that SlideCraft consistently produces high-quality, context-aware presentations tailored to diverse communication settings, with minimal human input and reliable adherence to enterprise constraints.
Murugan Sankaradas (presenting virtually) will present “TalentScout: Multimodal AI-Driven Expert Finding in Organizations” at the IEEE International Conference on Pervasive Intelligence and Computing (PICom2025) on Tuesday, October 21 (10:30am–12pm JST) | Monday, October 20 (9:30–11pm ET) in Hokkaido, Japan.
Kunal Rao (presenting virtually) will present “SlideCraft: Context-Aware Slides Generation Agent” at the IEEE International Conference on Pervasive Intelligence and Computing hashtag#PICom2025 on Tuesday, Oct 21 (10:30am–12pm JST) | Monday, Oct 20 (9:30–11pm ET) in Hokkaido, Japan. SlideCraft uses AI to automatically generate presentation slides from research content, making technical communication faster and context-aware for scientists and professionals.
Agentic AI systems rely on Large Language Models (LLMs) to execute complex tasks by invoking external functions. The efficiency of these systems depends on how well function execution is managed, especially under heterogeneous and high-variance workloads, where function execution times can range from milliseconds to several seconds. Traditional load-balancing techniques, such as round-robin, least-loaded, and Peak-EWMA (used in Linkerd), struggle in such settings: round-robin ignores load imbalance, least-loaded reacts slowly to rapid workload shifts, and Peak-EWMA relies on latency tracking, which is ineffective for workloads with high execution time variability. In this paper, we introduce Bifröst, a peer-to-peer load-balancing mechanism that distributes function requests based on real-time active request count rather than latency estimates. Instead of relying on centralized load-balancers or client-side decisions, Bifröst enables function-serving pods to dynamically distribute load by comparing queue lengths and offloading requests accordingly. This avoids unnecessary overhead while ensuring better responsiveness under high-variance workloads. Our evaluation on open-vocabulary object detection, multi-modal understanding, and code generation workloads shows that Bifröst improves function completion time by up to 20% when processing 13,700 requests from 137 AI agents on a 32-node Kubernetes cluster, outperforming both OpenFaaS and OpenFaaS with Linkerd. In an AI-driven insurance claims processing workflow, Bifröst achieves up to 25% faster execution.
In this paper, we propose a novel agentic AI system called XPF, which enables users to create “agents” using just natural language, where each agent is capable of executing complex, real-world business workflows in an accurate and reliable manner. XPF provides an interface to develop and iterate over the agent creation process and then deploy the agent in production when satisfactory results are produced consistently. The key components of XPF include: (a) planner, which leverages LLM to generate a step-by-step plan, which can further be edited by a human (b) compiler, which leverages LLM to compile the plan into a flow graph (c) executor, which handles distributed execution of the flow graph (using LLM, tools, RAG, etc.) on an underlying cluster and (d) verifier, which helps in verification of the output (through human generated tests or auto-generated tests using LLM). We develop five different agents using XPF and conduct experiments to evaluate one particular aspect i.e. difference in accuracy and reliability of the five agents with “human-generated” vs “auto-generated” plans. Our experiments show that we can get much more accurate and reliable response for a business workflow when step-by-step instructions (in natural language) are given by a human familiar with the workflow, rather than letting the LLM figure out the execution plan steps. In particular, we observe that “human-generated” plan almost always gives 100% accuracy whereas “auto-generated” plan almost never gives 100% accuracy. In terms of reliability, we observe through Rouge-L, Blue and Meteor scores, that the output from “human-generated” plan is much more reliable than “auto-generated” plan.
Latency-critical applications demand quick responses. Ideally, detailed insights are preferable for the best decision making and response actions. However, in situations when detailed insights cannot be provided quickly, even basic information goes a long way in tackling the situation effectively. For example, in marine security application, it is critical to immediately notify as soon as an unauthorized vessel is seen. Hence, timely response may be prioritized over the response based on entire details. To address such latency-critical situations, in this paper, we propose a novel system called DiCE-EC, which leverages LLM to generate distributed code with speculative execution on Edge (fast and simple response using resource constrained hardware) and Cloud (detailed response using powerful hardware, but may be fast or slow depending on network conditions). DiCE-EC breaks down application into smaller components and executes them asynchronously across the edge and cloud computing continuum. As network conditions vary, we show through real-world marine security application, that DiCE-EC is effective in dynamically choosing detailed insights from cloud when received within latency-constraint, or falling back to simple response from edge to guarantee timely alert delivery. Without such dynamic selection of response from edge or cloud, existing systems either always provide simple responses or drop alerts. We perform real network measurements in the Gulf of Pozzuoli in Naples, Italy along accessible areas (inland and in a Ferry) and generate 1 million realistic measurements across four inaccessible regions, and demonstrate that DiCE-EC never misses an alert, while baseline misses up to ?4% alerts with real data and up to ?1% (10,000 alerts) with generated data.
