Hi, I’m Priyanka Gupta, an AI Engineer based in Vadodara, Gujarat. I design and build end-to-end AI systems, from voice-to-voice assistants to secure data pipelines and cloud deployments. I enjoy turning complex problems into practical, scalable solutions that users can actually rely on. I thrive in collaborative, fast-paced teams and love exploring retrieval-augmented generation, vector stores, and automation on AWS. I’m always learning, prototyping, and refining models to deliver real business impact while keeping things robust and maintainable.…

Priyanka Gupta

Hi, I’m Priyanka Gupta, an AI Engineer based in Vadodara, Gujarat. I design and build end-to-end AI systems, from voice-to-voice assistants to secure data pipelines and cloud deployments. I enjoy turning complex problems into practical, scalable solutions that users can actually rely on. I thrive in collaborative, fast-paced teams and love exploring retrieval-augmented generation, vector stores, and automation on AWS. I’m always learning, prototyping, and refining models to deliver real business impact while keeping things robust and maintainable.…

Available to hire

Hi, I’m Priyanka Gupta, an AI Engineer based in Vadodara, Gujarat. I design and build end-to-end AI systems, from voice-to-voice assistants to secure data pipelines and cloud deployments. I enjoy turning complex problems into practical, scalable solutions that users can actually rely on.

I thrive in collaborative, fast-paced teams and love exploring retrieval-augmented generation, vector stores, and automation on AWS. I’m always learning, prototyping, and refining models to deliver real business impact while keeping things robust and maintainable.

See more

Skills

Experience Level

AWS
Expert
Python
Expert
Docker
Expert
PostgreSQL
Expert
SQL
Expert
GitHub Copilot
Expert

Language

English
Fluent

Work Experience

AI Engineer at Netweb Software
January 1, 2025 - Present
End-to-end AI system development: built an end-to-end voice-to-voice AI receptionist using OpenAI GPT-4 (STT, LLM, TTS), integrated with tool calling and secure PostgreSQL operations via APIs for dynamic scheduling and dialogue handling with barging-in functionality. Designed and developed a PostgreSQL-based tool interface leveraging MCP for live exchanges between local LLM agents and backend storage; implemented an MCP Inspector for real-time protocol validation, stateful session management, and debugging. Engineered a multi-layered architecture with vector stores and external data sources; automated LLM workflows with Weaviate, Apache Airflow, and n8n pipelines. Developed an agnostic SPA generator that interprets natural language prompts to produce front-end code, containerized with Docker, and auto-deployed to isolated environments for live previews and interaction. Led AWS-based RAG system development using Bedrock LLMs (Claude 3 Haiku), improving document retrieval accuracy and re
Machine Learning Intern at Solutions and Strategy
July 1, 2024 - September 1, 2024
Applied ML algorithms including LSTM and Auto ARIMA to build models predicting stock price movements with improved accuracy. Implemented and optimized models using Python ML libraries, performed extensive data analysis and feature engineering on historical stock data, and achieved reduced prediction errors and enhanced model performance.

Education

Bachelor of Engineering in Computer Engineering at Sardar Vallabhbhai Patel Institute of Technology, Vadodara, Gujarat
October 1, 2021 - May 1, 2025

Qualifications

British Airways Job Simulation on Forge
January 11, 2030 - April 2, 2026
Artificial Intelligence Foundations (LinkedIn)
January 11, 2030 - April 2, 2026
NN and CNN Essential Training (LinkedIn)
January 11, 2030 - April 2, 2026
Python for Data Science (LinkedIn)
January 11, 2030 - April 2, 2026

Industry Experience

Software & Internet, Computers & Electronics
    Enterprise RAG Chatbot for Internal Knowledge Search

    Developed a production-grade Retrieval-Augmented Generation (RAG) system that allows teams to query enterprise documents and receive accurate, context-aware answers. The solution included document ingestion, chunking, embeddings, vector search, prompt optimization, and API deployment using FastAPI, Qdrant, Weaviate, and AWS Bedrock. This significantly improved retrieval quality and reduced the time employees spent searching across documents manually.

    Business Agnostic AI Voice Receptionist for Appointment Booking

    Built an end-to-end voice-to-voice AI receptionist that can handle incoming customer calls, understand speech in real time, respond naturally, manage interruptions with barge-in support, and book appointments through secure scheduling APIs. The system integrates speech-to-text, LLM reasoning, text-to-speech, and PostgreSQL-backed workflows, making it suitable for clinics, salons, consultancies, and service businesses looking to automate front-desk operations.