Aniruddha Chaudhari

About Aniruddha Chaudhari

I’m a problem solver at heart, driven by a passion for transforming complex, messy data…

Experience

1. Technology Support Associate

   Heuristics Pharma Perceptions

   Jan 2025 — Present

   Engineered a model-serving infrastructure using Kubernetes and FastAPI, reducing model inference time by 30%, ensuring real-time predictions for high-volume analytical requests in business intelligence workflows.Developed an embeddings-based similarity model using FAISS and OpenAI embeddings, enhancing information retrieval accuracy by 15%, leading to faster access to relevant data for internal research teams.Built and fine-tuned NLP pipelines leveraging BERT models for automated entity recognition, increasing classification precision by 18%, enabling reliable categorization of unstructured datasets for regulatory analysis. Integrated a high-speed vector search pipeline with Neo4j and Elasticsearch, reducing data retrieval latency by 22%, allowing business users to query structured & unstructured data in under 2 seconds. Automated model monitoring and retraining workflows using MLflow & Prefect, reducing model drift by 19%, ensuring consistent model performance and reliability across evolving datasets.Orchestrated ETL workflows for unstructured data processing using Apache Airflow, automating data ingestion from APIs and internal sources, reducing manual processing efforts by 30%, and improving pipeline efficiency.