Rounak Saha

Machine Learning Engineer

Enquero•  July 2022 - Present•  Banglore

I am a Machine Learning Engineer at Enquero (a Genpact company), specializing in harnessing Large Language Models (LLMs), Generative AI, Deep Learning, MLOps, and NLP/NLU to drive innovation. My expertise spans advanced technologies such as Langchain, Vector DB (Chroma, Pinecone, and Weaviate), Llama Index, Fine Tuning (LORA & QLORA), and Knowledge Graph (Neo4J). My toolkit includes Azure ML Studio, AWS Sagemaker, Vertex AI, Kubeflow, PySpark, SQL, and Airflow tools that empower me to deploy successful solutions across AWS, Azure, and GCP, enhancing operational efficiency and decision-making. In my work, I leverage Transformer models from Hugging Face—BERT, Llama2/3, Falcon, and Mistral—alongside proprietary models like GPT-3.5, GPT-4, Claude 3 Haiku, Claude 3.5 Sonnet, Gemini, and Palm. This expertise allows me to specialize in LLMOps, RAG, Prompt Engineering, and Fine-tuning, ensuring cutting-edge Generative AI solutions.

NIT, Surathkal (National Institute of Technology)

Computer Science & Engineering, M.Tech•  September 2020 - July 2022•  CGPA: 9

I am pleased to have achieved an impressive percentile of 99.98% in the Graduate Aptitude Test in Engineering (GATE), which provided me with the opportunity to secure admission at the National Institute of Technology Karnataka (NITK). During my master's program at NITK, I had the privilege of engaging in a multidisciplinary journey, exploring various fields of study, including MLOps, Serverless Computing, Federated Learning, Time Series Analysis, and Hypothesis Testing. My dedication to advancing the frontiers of knowledge led to the authorship of four research papers during this period, with one of them achieving the honour of publication at the renowned IEEE BlackSeaCom 2022 conference in Sofia, Bulgaria. This achievement was made possible through the invaluable support of the Ministry of Human Resource Development (MHRD) Grant, underscoring the significance of the research contributions made during this time.