Hire Arun J. - LLM Engineer

* Working on AI memory infrastructure for Intuit's conversational AI platform, focused on production reliability, compliance, and adoption across internal teams. * Designed and built a Long-Term Memory (LTM) platform for AI agents, covering summarization, real-time memory, and semantic search — end to end, from architecture to production. * Platform handled 5.5M+ requests over 4 days during Tax Peak, peaking at 90 TPS with 99.999% availability. * Worked across security, privacy, PCI, and Responsible AI (RAI) review processes to ship AI capabilities in a regulated financial services environment. * Built MemoryLens, an observability and diagnostics layer for LTM systems, recognized as a Top 3 innovation at the VP of Engineering level. * Coordinated delivery across engineering, product, security, and compliance teams in a domain with evolving requirements and multiple external partners. * Mentored a junior engineer on the team through technical growth and ownership of complex systems.

* Spearheaded the development of Teladoc's first Generative AI-based project, creating high-quality clinical and chat summaries using techniques like prompt engineering, self-reflection and DSPy. Led the development of an AI chat assistant for chronic care members, incorporating Retrieval-Augmented Generation (RAG) and custom guardrails for patient safety. Lead engineer and significant contributor for developing and deploying Java, Scala ,Python-based APIs into production, successfully serving 1 million chronic care members. AI Skills & Technologies: High throughput asynchronous, non-blocking AI/ML APIs with Python, FastAPI, AsyncIO, Kafka. Experience with advanced LLM applications using Python, LangChain, Vector databases, Azure OpenAI / Open source Huggingface models, Kubernetes, Nemo Guardrails, MLFlow , RAGAS/LLM-As-Judge evaluation, Observability stacks. Knowledge of fine-tuning techniques, quantization, LLM serving libraries and optimizations. Backend Skills & Technologies: Scala(Play Framework), Java (Spring Boot), Python(FastAPI), Apache Kafka / Confluent cloud, Kafka Connectors, Kubernetes, Docker, Helm, MySQL, Redshift, Jenkins/CloudBees, Azure, Git , New Relic

* Lead engineer and significant contributor for developing and deploying Scala/Python-based services into production, successfully serving 1 million chronic care members. Led intiatives to identify, resolve, and remediate platform-level issues, such as the large-scale data inconsistencies across multiple services. Expert in troubleshooting and performance optimization, including resource contentions, SQL query performance problems, and bugs in code. Tech Stack : Scala, Python, Apache Kafka / Confluent cloud, Kafka Connectors, Kubernetes, Helm, MySQL, Redshift, Jenkins/CloudBees, Azure, Git , New Relic