Sukumar Dhangar

• Senior AI Engineer with expertise in building Generative AI solutions and fine-tuning large language models (LLMs) such as GPT, Llama and GPT-Real time. Specialized in developing scalable AI applications, optimizing model performance, and deploying NLP-based systems. Skilled in end-to-end AI workflows, from model training to production deployment on cloud platforms.

Projects:

1. Generative AI-Powered Voice Bot for Call Center Cost Reduction.

Created a voice bot using Speech-to-Text (STT), Text-to-Speech (TTS), and Large Language Models (LLM) to sell personal loans.

Impact

Managed 300,000 calls daily, driving monthly personal loan sales of ₹300 crore.

Technologies:

Prompt Engineering, Python, STT, Text-to-Speech (TTS), Rest API, Python, LLM, JSON, Flask Server, Azure OpenAI, and LangChain, Linux.

2. Generative AI-Powered OCR Solution for Invoice Processing.

Developed an OCR system powered by generative AI for accurate and cost-effective invoice processing.

Impact:

Process the 100,000 invoices daily, reduce manual processing time, and streamline the overall invoice management workflow.

Technologies: LLM, Python, Automation, Azure OpenAI, React, Fast, JSON, REST API.

3. Generative AI-Powered Service Chat Solution.

Developed a service chatbot using Large Language Models (LLMs), Azure Cognitive Services (ACS), and Python to handle customer queries efficiently.

Impact: Successfully resolved an average of 80,000 customer queries per month.

Significantly reduced operational costs by minimizing manual intervention.

Technology:
Azure Cognitive Services (ACS), Large Language Models (LLM), Python , Prompt Engineering, Database, Rest Api, RAG, Chroma-DB