Machine Learning Lead at Adsys a med-tech startup based in Jaipur, India.
In my prior role as ML-OPS Head at Appyhigh Technologies, a mobile internet startup, I specialized in deploying pipelines for models using frameworks like TensorFlow Serving, TorchServe, and Nvidia Triton Server.
My expertise spans in deploying and fine-tuning Large Language Models (LLMs) and other cutting-edge Generative AI models, as well as managing efficient data pipelines.
Furthermore, I bring valuable experience in constructing Elastic Search and Apache Airflow pipelines to enhance organizational workflows.
In addition to my professional experience, I hold a Bachelor's degree in Electronics and Communication Engineering from The LNM Institute of Information Technology (LNMIIT).
Phot.AI, an innovative ecosystem developed by Appyhigh Technologies, seamlessly integrates Generative AI Tools, APIs, and Plugins for advanced AI-powered photo editing and visual design.
As the ML-Ops head of this project, my responsibilities included deploying cutting-edge generative AI models inference pipelines through tools like Kubeflow, while also optimizing server costs to enhance efficiency.
A RAG(Retrieval Augmented Generation) Agent created using Langchain and Open-Sourced LLMs and hosted on Huggingface Spaces.
Continuning from the work of clement.
This chatbot is adept at formulating answers based on the analysis of content from not only uploaded PDFs but also from Hugging Face blog posts.
Fine-Tuned a 7billion parameter model using huggingface on a 24GB VRAM A10g. The process involved creation of synthetic dataset using GPT4 APIs, which was cleaned before undergoing Parameter-Efficient Fine-Tuning (PEFT) on our systems. This process leveraged advanced quantization methods provided from bits and bytes and Axotol.
The Model will be hosted using TGI framework by huggingface.