I am Ghazala Arshad, an AI Agent Developer and LLM Engineer based in Hyderabad, Pakistan. I specialize in Retrieval-Augmented Generation (RAG) and multi-agent systems, building low-latency AI agents using frameworks like CrewAI and LangChain. I recently designed and deployed a bilingual (Hindi & English) RAG-based AI agent for a remote project, with sub-second inference latency achieved by integrating Groq Llama3-70B and robust vector stores. I ranked in the Top 1% globally in Google's Prompt-to-Agent Engineering Challenges 2025 and published a complete RAG + Multi-Agent system demo on GitHub. I am passionate about delivering efficient, reliable AI solutions for clients worldwide.

Ghazala Arshad

I am Ghazala Arshad, an AI Agent Developer and LLM Engineer based in Hyderabad, Pakistan. I specialize in Retrieval-Augmented Generation (RAG) and multi-agent systems, building low-latency AI agents using frameworks like CrewAI and LangChain. I recently designed and deployed a bilingual (Hindi & English) RAG-based AI agent for a remote project, with sub-second inference latency achieved by integrating Groq Llama3-70B and robust vector stores. I ranked in the Top 1% globally in Google's Prompt-to-Agent Engineering Challenges 2025 and published a complete RAG + Multi-Agent system demo on GitHub. I am passionate about delivering efficient, reliable AI solutions for clients worldwide.

Available to hire

I am Ghazala Arshad, an AI Agent Developer and LLM Engineer based in Hyderabad, Pakistan. I specialize in Retrieval-Augmented Generation (RAG) and multi-agent systems, building low-latency AI agents using frameworks like CrewAI and LangChain. I recently designed and deployed a bilingual (Hindi & English) RAG-based AI agent for a remote project, with sub-second inference latency achieved by integrating Groq Llama3-70B and robust vector stores.

I ranked in the Top 1% globally in Google’s Prompt-to-Agent Engineering Challenges 2025 and published a complete RAG + Multi-Agent system demo on GitHub. I am passionate about delivering efficient, reliable AI solutions for clients worldwide.

See more

Experience Level

Expert
Expert
Intermediate
Intermediate
Intermediate
Intermediate

Language

Urdu
Fluent
English
Fluent

Work Experience

Senior AI Agent Developer
January 1, 2025 - Present
Designed and deployed bilingual (Hindi & English) RAG-based AI agent using CrewAI and LangChain; integrated Groq Llama3-70B achieving sub-second inference latency; implemented ChromaDB vector storage with hallucination prevention guardrails; GitHub: https://github.com/Ghazala001/AI-Agent-RAG-Demo
Freelance AI Agent Developer & Prompt Engineer at Freelancer.com
January 1, 2023 - Present
Designed and deployed bilingual RAG-based AI agent using CrewAI and LangChain; integrated Groq Llama3-70B achieving sub-second inference latency; implemented ChromaDB vector storage with hallucination prevention guardrails; GitHub: https://github.com/Ghazala001/AI-Agent-RAG-Demo

Education

Bachelor’s Degree at University of Sindh, Jamshoro
January 1, 2019 - January 1, 2022
Bachelor’s Degree at University of Sindh, Jamshoro
January 1, 2019 - December 31, 2022

Qualifications

Ranked Top 1% globally in Google Prompt-to-Agent Engineering Challenges 2025
January 1, 2025 - December 24, 2025
Completed Creative, Analytical, and Refactoring challenge tracks
January 1, 2025 - December 24, 2025
Built and deployed a full RAG + Multi-Agent system within 48 hours
January 1, 2025 - December 24, 2025
Google Prompt-to-Agent Engineering Challenges 2025 - Top 1% Global
January 1, 2025 - December 26, 2025

Industry Experience

Computers & Electronics, Software & Internet, Media & Entertainment, Professional Services
    paper Ai Agent with RAG and Multi-Agent Workflow)

    Designed and deployed a complete bilingual (Hindi & English) Retrieval-Augmented Generation (RAG) based multi-agent AI system as part of Google’s Prompt-to-Agent Engineering Challenges 2025.

    Key features:

    • Built using CrewAI and LangChain frameworks
    • Integrated Groq Llama3-70B for sub-second inference latency
    • Implemented ChromaDB vector storage with hallucination prevention guardrails
    • Multi-agent workflow with planning, reflection, and tool orchestration

    This project ranked me in the Top 1% globally in the challenge, completed in just 48 hours.

    Live Demo & Code: https://www.twine.net/signin

    #AI #LLM #RAG #MultiAgent #LangChain #CrewAI #Groq #PromptEngineering #Python