Arpit Aggarwal

Built and led a high-performing AI organization, scaling to 12 engineers, researchers, and scientists focused on deep learning, computer vision, and applied machine learning in telecommunications. Spearheaded the integration of GenAI, LLMs, and RAG technologies to drive innovation across network intelligence, marketing, and customer engagement.

• Architected predictive maintenance and remote sensing solutions using satellite imagery and deep learning for telecom infrastructure (towers, fiber, base stations), reducing downtime and OPEX by 25%.
• Pioneered GenAI use cases in telecom operations and marketing, including LLM-powered tools for:
  Incident summarization and shift-handover automation
  Intelligent assistants for field technicians and NOC engineers
  Real-time campaign content generation and personalization
• Developed LLM-based marketing content engines that generated localized campaign copy, offer descriptions, and A/B testing variants—reducing creative cycle time by 60% and enabling hyper-personalized, geo-targeted communications at scale.
• Implemented RAG systems that ingested market research, competitor intelligence, and internal campaign performance data to produce auto-generated campaign briefs, marketing playbooks, and executive insights reports.
• Designed customer engagement copilots using fine-tuned LLMs that analyzed interaction data, tickets, and product usage to suggest next-best actions, upsell strategies, and personalized messaging—resulting in a 15% increase in conversion across digital channels.
• Led ML initiatives for churn prediction, CLTV forecasting, and segmentation, enhanced with GenAI for sentiment extraction from voice calls, chat logs, and survey data.
• Embedded LLMs into BSS and CRM systems to assist support agents with summarizing customer history, drafting real-time responses, and surfacing relevant offers dynamically during interactions.
• Leveraged geospatial AI + RAG frameworks to support 5G rollout planning, using structured network data and unstructured regulatory/market intelligence to recommend optimal deployment zones.
• Advanced multi-modal AI systems combining RF signal data, imagery, and text for comprehensive anomaly detection and root-cause narrative generation in hybrid network environments.
• Worked cross-functionally with the CTO, CMO, and product teams to align AI strategy with revenue, retention, and operational KPIs, contributing to enterprise-wide digital transformation and customer-centric innovation.
• Published research and delivered thought leadership on the application of GenAI, deep learning, and statistical modeling in telecom network management and customer analytics.
• Advocated and implemented Responsible AI governance, including fairness audits for marketing personalization models, transparency in customer-facing LLMs, and regulatory alignment in AI deployments across critical systems.

• Spearheaded AI strategy and innovation for the Emerging Business Unit (EBU), delivering cutting-edge solutions across utilities, energy, transport, travel, and operational technology sectors.
• Introduced scalable AI platforms integrating edge computing, MLOps pipelines, and secure data infrastructures for critical OT environments.
• Collaborated with business leaders, product managers, and technical stakeholders to translate emerging market needs into high-impact AI solutions, accelerating new revenue streams for the EBU.
• Developed intelligent transportation solutions, leveraging computer vision and geospatial AI to optimize traffic flow, monitor infrastructure health, and improve safety standards.
• Architected AI systems for operational technology environments, enhancing security, anomaly detection, and resilience of critical infrastructure through multimodal data fusion and real-time analytics.
• Partnered with product, engineering, and industry-specific business units to co-create AI solutions tailored for enterprise clients, significantly accelerating time-to-market for innovative offerings.
• Built strategic alliances with leading universities, research labs, and technology providers to strengthen the AI innovation pipeline for EBU sectors.
• Advised senior leadership on AI investments, market positioning, and emerging technology trends, directly contributing to business growth and expansion into new verticals.
• Advocated for responsible AI development, embedding ethical guidelines, fairness auditing, and explainability frameworks across all EBU AI projects.
• Championed ethical AI frameworks, particularly around security, resilience, and compliance in highly regulated sectors.

• Led AI innovation across pharmaceutical R&D consulting projects, transforming clinical trial design, molecule discovery, and pharmacometric modeling through advanced deep learning and generative AI techniques.
• Directed simulation-based studies for clinical trials and randomized controlled trials (RCTs), employing AI-driven trial simulation engines to optimize patient recruitment strategies, dose selection, and endpoint prediction.
• Developed generative AI models and Variational Autoencoders (VAEs) to accelerate de novo molecule generation, virtual screening, and predictive modeling for drug discovery pipelines, reducing candidate identification timelines by 30%.
• Built Bayesian PPK-PD models integrating real-world evidence and trial data, enabling personalized dose optimization and improving therapeutic index predictions for late-phase clinical trials.
• Applied Markov Chain modeling and Monte Carlo simulations to support pharma clients in risk assessment, trial outcome prediction, and regulatory submission planning.
• Spearheaded AI use cases including synthetic patient population generation, biomarker discovery, toxicity prediction, and adaptive trial design using deep learning frameworks.
• Collaborated with biostatisticians, clinical scientists, and regulatory experts to embed AI-enhanced modeling in submission packages for FDA/EMA approvals.
• Architected custom AI platforms that integrated imaging, genomic, and clinical data, enabling multi-modal analysis for precision medicine initiatives.
• Provided AI strategy consulting to top pharma companies, helping shape digital R&D roadmaps and drive AI-first transformation in clinical operations.
• Advocated for ethical AI adoption in pharma, ensuring model transparency, explainability, and compliance with GxP and regulatory standards.

• Led AI initiatives across heavy manufacturing sectors, delivering transformative solutions in steel-making, boiler operations, HVAC systems, industrial automation, and supply chain optimization.
• Developed and deployed deep learning models for defect detection in steel production lines using multispectral and thermal imagery, improving quality control accuracy.
• Architected predictive maintenance platforms using sensor fusion (vibration, thermal, acoustic data) and time-series forecasting to extend asset lifespans and reduce unplanned downtimes across boilers, compressors, and HVAC units.
• Built digital twin models combining physics-based simulations and machine learning to optimize boiler efficiency, emissions control, and energy management in industrial plants.
• Applied reinforcement learning and optimization algorithms to dynamically manage supply chain networks, inventory planning, and production scheduling, achieving 30% cost and time savings.
• Designed AI-based anomaly detection systems for industrial IoT data streams, enabling early fault detection and proactive interventions in critical operations.
• Led use of computer vision for automated inspection of welds, surface finishes, and structural components in heavy manufacturing, reducing manual inspection overheads.
• Partnered with engineering, plant operations, and quality teams to embed AI into daily operations, accelerating digital transformation and smart factory initiatives.
• Provided strategic consulting on AI roadmap creation, technology selection (edge AI, industrial MLOps), and cybersecurity for AI-enabled manufacturing environments.
• Advocated responsible AI deployment, emphasizing explainability, robustness, and safety in mission-critical industrial systems.

Led strategic AI, ML, Econometrics, and GenAI engagements across multiple sectors including retail, supply chain, logistics, oil & gas, food & beverage, and marketing. Bridged business strategy with advanced analytics to deliver high-impact solutions and guide C-level decision-making.

Key Achievements:

• Delivered $10M+ in client value through AI-powered optimization, demand forecasting, and revenue growth initiatives.
• Developed and deployed 10+ enterprise-grade GenAI solutions, including LLM-based assistants for supply chain, marketing, and technical documentation.
• Reduced supply chain forecasting errors by 20–30% using time-series ML and econometric models.
• Led AI/ML strategy engagements for clients in retail, transportation, and energy, aligning analytics roadmaps with operational goals.

Representative Use Cases:

• Retail: Built a dynamic pricing engine using ML and price elasticity modeling; deployed GenAI-powered retail trend summarizer.
• Supply Chain: Designed inventory optimization models and LLM assistants for planner decision support.
• Logistics: Implemented fleet routing optimization via reinforcement learning; econometric cost driver analysis for fuel efficiency.
• Oil & Gas: Predictive maintenance with sensor data; econometric models for price sensitivity under volatility; LLMs for summarizing technical SOPs.
• Marketing: Marketing mix modeling with causal ML; campaign brief generator using persona-driven GenAI.
• Food & Beverage: Customer churn prediction and segmentation; GenAI-driven product innovation assistant.

Led advanced modeling and simulation initiatives to support risk assessment and pricing strategies for natural catastrophe events in the insurance and reinsurance sector. Specialized in large-scale mathematical modeling, spatial-temporal simulations, and AI/ML applications for satellite and sensor data analysis.

Key Achievements:

• Designed and calibrated large-scale catastrophe risk models for hurricanes, earthquakes, and floods using simulation frameworks, probabilistic modeling, and optimization techniques.
• Developed Exceedance Probability (EP) curves and loss distribution models to inform risk pricing, capital allocation, and reinsurance structuring.
• Applied Genetic Algorithms and Bayesian Optimization to calibrate stochastic hazard models using historical event data and real-time observations.
• Delivered end-to-end ML pipelines for automated hazard detection and severity estimation from satellite imagery, LiDAR, IoT, and radar sensor feeds.

Representative Projects:

• Natural Hazard Mapping: Built ML-based classifiers and segmentation models (CNN, U-Net) for flood extent and wildfire boundaries using multispectral satellite data (e.g., Sentinel, Landsat).
• Seismic Risk Modeling: Created Monte Carlo simulation engines to model ground motion propagation and asset vulnerability; improved event correlation logic using copulas and spatial dependency modeling.
• Climate Impact Modeling: Integrated CMIP6 climate projections into catastrophe risk models to support climate-adjusted underwriting and strategic portfolio steering.
• Exposure & Asset Data Enrichment: Used geospatial AI and computer vision to extract and classify structural features from aerial imagery to enhance underwriting data.
• Model Validation & Governance: Implemented rigorous model validation workflows, uncertainty quantification, and regulatory reporting aligned with NHFL NAHU and FEMA guidelines.