Rohanth Gundu

Jobs 0% verified

AI/ML Engineer

Perplexity

Jan 2024 - Current (2 years 5 months)

Architected retrieval-augmented generation (RAG) pipelines processing 1M+ monthly queries, integrating semantic search and LLM-driven contextual retrieval to boost accuracy and responsiveness by 40%.Developed real-time data ingestion and feature extraction pipelines using Kafka, gRPC, and Spark to support continuous model training and inference under sub-60 ms latency SLAS.Built and fine-tuned custom embedding models with PyTorch and TensorFlow, storing high-dimensional vectors in Pinecone for scalable retrieval and personalized recommendations.Designed GPU-optimized inference microservices on Kubernetes (AWS, Terraform, Helm), automating scaling for large transformer-based models in production.Implemented event-driven ML workflows via NATS
AI/ML Engineer

Meta

Jun 2019 - Oct 2023 (4 years 5 months)

Designed and deployed large-scale ML models for recommendation systems, NLP, and computer vision pipelines, improving user engagement by 18%.Built semantic retrieval systems using Hugging Face Transformers, LangChain, and FAISS vector databases, reducing search latency by 45%.Led end-to-end LLM-based question-answering pipelines, integrating prompt engineering, retrieval augmentation, and deployment for internal knowledge access.Developed MLOps pipelines using Airflow, MLflow, and Kubernetes (EKS), integrating feature engineering, hyperparameter tuning, and model monitoring.Containerized ML workflows with Docker and deployed on cloud infrastructure (AWS, GCP), enabling horizontal auto-scaling and high availability.Engineered ETL pipelines a

Education 0% verified

About

Detail

Timeline

Résumé