kapil s.

I am a Senior Data & AI Engineer with more than 15+ years of experience. I build data platforms, data pipelines, machine learning solutions and AI or Generative AI systems for both startups and large companies. My work covers the data and AI life cycle. I handle data ingestion, ETL or ELT data modeling, warehousing, machine learning, LLM applications, Retrieval Augmented Generation (RAG) systems, AI agents, analytics and production deployment. I use AWS, Azure, GCP, Databricks, Snowflake and the newest AI and machine‑learning tools. Recent Projects - Built an inbound voice system with Twilio, OpenAI Whisper and GPT‑4 that takes voice input in real time and routes calls. - Created LLM and RAG systems for company knowledge bases, documents and business data. - Developed an AI platform that integrates HL7 FHIR R4 with Epic, Cerner and Athenahealth. - Cut LLM inference costs by twenty‑five percent for a high‑volume voice AI product by improving OpenAI Realtime API use and token handling. - Designed scalable data pipelines and platforms using Apache Spark, PySpark, Airflow, dbt, Databricks, Snowflake, Kafka and cloud services. - Built machine‑learning pipelines and predictive models with PyTorch, TensorFlow, scikit‑learn and MLflow. - Built Power BI data models and reporting for healthcare and finance. Data Engineering I have experience building ETL or ELT pipelines, data warehouses, lakehouse platforms, batch and real‑time pipelines and data models. I use Python, SQL, PySpark, Apache Spark, Airflow, dbt, Kafka, Kinesis, Fivetran, Databricks, Delta Lake, Snowflake, BigQuery, Redshift and AWS Glue. AI, Generative AI & Machine Learning I create production AI solutions that use LLMs, RAG, AI agents, AI assistants, natural‑language processing, embeddings, vector search, machine learning and ML pipelines. My hands‑on tools include OpenAI GPT‑4 or GPT‑4o, Claude, Gemini, LangChain, LlamaIndex, Pinecone, Weaviate, pgvector, PyTorch, TensorFlow, scikit‑learn and MLflow. I focus on more than connecting an application to an LLM. I engineer the surrounding parts: data preparation, retrieval, evaluation, prompt design, model integration, cost control, scaling, monitoring and production reliability. Infrastructure I work on AWS, Microsoft Azure and Google Cloud. I use services such as AWS Glue, EMR, Lambda, S3, Redshift, Athena, SageMaker, GCP BigQuery, Dataflow, Vertex AI, Azure Data Factory, Azure Synapse and Azure ML. I also use Docker, Kubernetes, Terraform, GitHub Actions and CI/CD to deploy and manage production data and AI workloads. Healthcare, Security & Compliance I work with healthcare and sensitive data. I handle HL7 FHIR, electronic health records, HIPAA, GDPR, SOC 2 personal data protection, role‑based access control, encryption, data masking, access controls and data lineage. In enterprise settings I treat security, privacy, governance and compliance as part of the design from day one instead of adding them later. How I Work I begin by learning the business problem the current systems and the data architecture. Then I choose the solution. My goal is to build systems that're reliable, scalable, maintainable, secure and useful for the business. Whether you need a data platform, ETL pipeline, lakehouse, machine‑learning solution, LLM or RAG application AI agent, analytics platform or AI infrastructure I can own the project from architecture through deployment and ongoing improvement. I have delivered projects for startups and large companies, in the United States, Europe and Asia in healthcare, finance, SaaS, e‑commerce and other data‑heavy industries. If you need a Senior Data & AI Engineer who can connect your data, AI and cloud services into a solution I would be glad to talk about your project.