PREM JHA

I specialize in building and exploring Agentic AI systems that automate real-world workflows. My focus is on developing AI agents that can understand objectives, reason through tasks, use tools, maintain state, execute actions, validate results, and handle failures rather than simply generating text.

My key project is an AI Career Assistant, where I combine LLMs, Python, APIs, databases, resume processing, browser automation, notifications, and workflow management. The system is designed to analyze job requirements, understand resume information, tailor resumes for specific roles, track applications, and support browser-based job application workflows.
I have hands-on experience with Gemini API, LangChain, Python, SQL, Machine Learning, data analysis, APIs, and browser automation. I focus on separating LLM reasoning from deterministic application logic so that important operations such as validation, database updates, file handling, and browser actions remain controlled and reliable.
What sets me apart is that I don't view an AI agent as simply an LLM with a prompt. I think about the complete lifecycle: Understand → Plan → Act → Observe → Validate → Recover. I also emphasize state management, error handling, modular architecture, and human-in-the-loop control for actions that require user confirmation.
My goal is to move beyond basic chatbots and build practical AI agents capable of automating complex, multi-step workflows while remaining reliable, explainable, and useful in real-world applications.