Ansh Rawat

About Ansh Rawat

About MeI am an AI Engineer and technology enthusiast with experience in building LLM-powered applications, chatbots, and retrieval-augmented generation (RAG) pipelines. My journey started as a Gold Microsoft Learn Student Ambassador, where I built thriving student communities, mentored peers, and honed my leadership, collaboration, and problem-solving skills.I hold a B.Tech in Computer Science from Chandigarh University, where I built strong foundations in data, software development, and analytical thinking. Since then, I’ve expanded my expertise into applied AI, working with tools like LangChain, Hugging Face, FAISS, Cassandra Vector DB, Whisper AI, and Gradio to develop intelligent, real-world prototypes. What I bring to the table:Hands-on experience with chatbots, embeddings, and multi-modal AI systemsStrong background in Python, LLM integration, and AI deployment workflowsProven ability to bridge technology and people, from coding scalable AI solutions to leading collaborative projectsPassion for research, experimentation, and turning cutting-edge AI into practical solutions My mission is to empower people through technology—whether by building intelligent AI systems, supporting innovation communities, or helping others unlock their potential with the right tools.Always curious. Always learning. Always building.

Experience

1. Associate Software Developer

   Embitel Technologies

   Jun 2025 — Present · Bengaluru, IN

   I am working on full-stack conversational AI system that enables natural and seamless communication between users and an AI assistant through both text and voice. The goal of this project is to create an assistant that not only chats intelligently but also understands and responds in multiple languages, making it more inclusive and user-friendly.What the project does* Text & Voice Interaction: Users can either type or speak their queries, and the assistant responds in both text and spoken form.* Real-Time Speech Recognition (STT): Converts user’s speech into accurate text instantly.* AI-Powered Conversations: The backend integrates advanced LLMs to generate contextual and human-like responses.* Multilingual Support: The assistant can operate in English and Indian regional languages, bridging the gap for non-English speakers.* Transliteration Layer: Handles code-mixed inputs (like Hinglish) and converts them into natural, phonetically correct language text for smooth processing.* Voice Output (TTS): Responses are synthesized into natural-sounding speech, allowing the assistant to “talk back” in the chosen language.* Streaming Audio Experience: Even though some providers return audio in chunks, the system streams it intelligently so users hear responses smoothly without long waits.* Using AI agents and function calling to perform certain tasks like extracting discounts, cart details etc * Provided toggle switch to toggle between multiple languages and llm models to test.Tech Stack* Backend: Python (Flask), APIs for Speech-to-Text, Text-to-Speech, and Conversational AI* Frontend: JavaScript, TailwindCSS, Web Audio API, MediaRecorder API for real-time voice handling* Integrations: OpenAI, Azure Cognitive Services, Sarvam.ai (for multilingual capabilities)* Other Tools: LocalStorage (to persist chat sessions).env for secure config management

Education

• Bhartiya Vidya Bhavan Vidyashram Jaipur

  High School Diploma, High School/Secondary Diplomas and Certificates

  2005 — 2020

• CHANDIGARH UNIVERSITY

  Bachelor of Technology - BTech, Computer Science

  2020 — 2024