Anandhu Krishna

– Developed and integrated a benchmarking framework to analyze the quality of AI-generated data. – Improved the quality of AI services across customer interactions via audio, chat, and email. – Integrated various AI services into an existing quality monitoring SaaS application. – Contributed to the development of the application LLM augmentistic, enhancing overall functionality. • Imrpveing the client call auido transcription using different speech to text methods

– Developed a production ready module for automatic parsing and modeling of complex insurance contracts, forming the entry point for AXA's AI powered Computable Contracts platform. – Benchmarked and evaluated over 25 document parsers, leading to the adoption of a state-of-the-art multimodal LLM approach GPT-40 for robust handling of unstructured legal documents and nested tables. – Refactored the workflow into a modular codebase using Clean Architecture, enabling maintainable, testable, and scalable development for future enterprise deployment. – Used graph databases (kuzu, neo4j) to map out relationships in contracts, making it easier to run smart searches and analysis. – Engineered a comprehensive evaluation pipeline and custom metrics (NID Score, TEDS Score); documented all research and experiments in a GitHub wiki to ensure reproducibility and support cross-team learning. – Containerized all project components for reliable deployment and introduced observability with Langfuse to optimize LLM prompt engineering, supported by automated CICD pipelines and team-based code reviews. – Collaborated daily in an agile team environment, delivered regular demos, and contributed to rigorous documentation and knowledge transfer practices for project continuity.

– Developed and deployed advanced machine learning models tailored for the telecommunications industry, leading to impactful business results. – Churn Prediction Model: Developed a model that cut customer attrition by 15%, enabling proactive retention by identifying early warning signs. – Customer Engagement Campaigns: Created a predictive model that boosted engagement by 25% and raised order values by 10% through intelligent recommendations. – Scalable Model Deployment: Deployed ML models on Linux servers, ensuring scalability, optimal performance, and seamless integration with telecom systems. – Cross-functional Collaboration: Collaborated with global clients and teams to align analytical solutions with business goals, enhancing satisfaction and retention.