Responsibilities include:

• Analyze current and past employee data
• Look at employee performance and identify trends
• Prepare reports on the above information and communicate the insights of these reports to the HR supervisor
• Suggest improvements based on the above information

• Leading the development of AI agent workflows to automate the analysis and interpretation of security threats in electronic messages.
• Building and optimizing Python APIs (FastAPI) to deliver real-time security assessments to browser and email plugins.
• Engineering prompt strategies to improve AI accuracy in identifying phishing tactics, working towards a 90% success rate for actionable user advice.
• Technologies: Python, FastAPI, AWS, LLM/Prompt Engineering, Agno

• Built a local LLM-powered CLI assistant that converts natural language prompts into structured JSON actions for execution.
• Integrated Ollama as the inference backend with a regex fallback for offline reliability.
• Designed a modular skill system for extendable capabilities (e.g., file management, analysis).
• Added safety gating for destructive actions and planned voice/UI extensions

Developing a Threat Intelligence Dashboard that aggregates, analyzes, and visualizes cybersecurity data from multiple trusted sources. The system integrates APIs from VirusTotal and Google OSV to collect real-time information on malware, indicators of compromise (IOCs), and known software vulnerabilities.

Built an interactive dashboard analyzing over 10 years of stock data from 30 major U.S. companies, sourced from Kaggle. Implemented a data pipeline in Python (pandas, NumPy) to clean and process thousands of records, then visualized trends using Matplotlib/Plotly. Key features include filtering by company, timeframe, and metrics such as volatility and growth. The dashboard enables side-by-side performance comparison and highlights anomalous patterns across market cycles. Structured for scalability with modular scripts and clear project organization.

Developed a web application in a 24-hour hackathon aimed at helping residents assess their fire-hazard safety readiness. As a complete beginner to web development at the time, I worked with my team to design a simple yet functional interface where users could input household data (e.g., fire alarms, escape routes, extinguisher availability). The app assigned a readiness score and generated a detailed safety report based on weighted criteria.