Natasha Mittal

About Natasha Mittal

AI / ML Engineer with hands-on experience in building, integrating, and deploying machine learning and Generative AI solutions using Python. Currently working as an AI/ML Developer, focused on developing AI-powered applications involving large language models (LLMs), NLP pipelines, and data-driven systems.Experienced in machine learning, natural language processing (NLP), and end-to-end ML workflows, including data preprocessing, feature engineering, model training, evaluation, and optimization. Actively working with Generative AI concepts such as prompt engineering, embeddings, retrieval-augmented generation (RAG), and LLM orchestration using frameworks like LangChain and Hugging Face Transformers.Skilled in developing AI solutions such as document Q&A systems, chatbots, text summarization workflows, and ML-based classification models. Familiar with vector databases and semantic search techniques using FAISS, as well as API-based integrations with AI model providers. Strong understanding of model performance optimization, latency, and cost considerations.Holds a B.S. in Data Science and Applications from IIT Madras, with a strong foundation in statistics, machine learning algorithms, and applied data analytics. Continuously upskilling in Generative AI, LLM architectures, MLOps fundamentals, and cloud-based AI deployment (AWS / GCP). Open to AI Engineer, Machine Learning Engineer, and Generative AI Engineer opportunities.AI / ML Engineer | Generative AI | LLMs | RAG | LangChain | NLP | Python | Hugging Face | Vector Databases | Chatbots

Experience

1. Ai Ml Developer

   Dotsquares

   Oct 2025 — Present · Jaipur, IN

   Architected and deployed production-grade AI Agents and multi-modal RAG pipelines using FastAPI, LangChain, LangGraph, Llama-3, OpenAI/Groq APIs, ChromaDB, and PostgreSQL, enabling real-time document Q&A, semantic search, and conversational workflows. • Handled an LLM-powered EdTech Customer Support AI system (Next.js, Node.js, Vercel AI SDK) with intent detection, role-based orchestration, embeddings, and vector retrieval, improving issue resolution rate by 40% and reducing support tickets via workflow automation. • Developed a stateful conversational AI agent with persistent memory, tool-calling, real-time token streaming (SSE), and GPT-4o integration, optimizing context retention and response accuracy in multi-turn interactions. • Engineered scalable backend systems using Python, FastAPI, SQLAlchemy, Pydantic, REST APIs, integrated Whisper (ASR) for audio ingestion, and implemented production-ready LLM infrastructure with Git-based CI and containerized deployment

Education

• Indian Institute of Technology, Madras

  Bachelor of Science - BS, Data science and Programming

  2020 — 2025

• S S Jain Subodh PG Mahila College, Jaipur

  BCA, Computer and Information Sciences and Support Services