Projects & experience.
A closer look at the data, security, and Python work behind the résumé. Every project links to the source on GitHub.
RAGmind · Agentic AI Retrieval System
An agentic retrieval-augmented generation (RAG) system that answers cybersecurity questions from real reference material. It does not just retrieve and answer, it judges whether it found enough, retries with a better query when it has not, cites every source, and refuses to guess when the answer is not in its documents. Built, evaluated, and deployed end to end.
- Built hybrid search that blends semantic embeddings with keyword scoring, so it handles both paraphrased questions and exact terms like specific vulnerability identifiers.
- Designed an agentic self-correction loop: the system grades its own retrieval and rewrites the query to search again when the first attempt falls short.
- Engineered grounded answers with inline citations and a deliberate refusal behavior to avoid hallucinating beyond the source material.
- Wrote a built-in evaluation harness that measures retrieval accuracy, answer quality, and how often the system correctly refuses unanswerable questions.
Password & Breach Risk Analyzer
A live web app that scores any password against real-world weak-pattern analysis and a list of commonly leaked passwords, then estimates how quickly it could be cracked. Built with Python and deployed with Streamlit.
- Analyzed a dataset of commonly leaked passwords to find the patterns that make passwords weak, then visualized the findings.
- Built a scoring engine that checks length, character variety, breach-list matches, and obvious weak patterns.
- Designed the tool to analyze input locally without storing it, a deliberate security-conscious choice.
Loan & Branch Performance Analysis
Selected for the ALPHA LIFT Financial Technology Summer Mentorship, sponsored by Bank of America, where the focus was turning raw banking data into clear performance stories.
- Analyzed loan and branch datasets in Excel using VLOOKUP and CUMIPMT to evaluate performance trends.
- Built visualizations that ranked branches and identified regional and seasonal patterns.
- Delivered data driven insights on risk exposure and growth opportunities to program mentors.
Cloud Computing & Security Research
An independent research project examining how modern cloud systems are built and where they are most exposed.
- Researched common cloud architectures and the attack vectors that target them.
- Authored a technical paper documenting risks alongside concrete mitigation strategies.
- Connected security concepts from coursework to real architectural trade offs.
Python Data Analysis Project
A data analysis project built in Python, applying core analysis workflows with the libraries used across the data field. Source published on GitHub.
- Applied Python data tooling such as Pandas and NumPy to work through a dataset.
- Practiced a repeatable load, clean, and analyze workflow.
Python Practice Suite & Exercises
A collection of small programs written to strengthen core programming fundamentals and problem solving.
- Built logic for even and odd detection, vowel and consonant checks, days in a month, and letter grade conversion.
- Created scripts for countdown timers, minimum and maximum input, filtering even numbers from lists, and computing grade averages.
- Wrote a temperature conversion tool that takes Celsius input and returns Fahrenheit with formatted output.
- Reinforced conditionals, control flow, loops, functions, and input validation.
Patient Service Coordinator, Ritual Dental
A part time role in New York, NY that turned into a chance to improve the data and payment workflows behind the front desk.
- Analyzed payment workflows and proposed automation for transaction tracking.
- Improved efficiency through secure payment link recommendations.
- Maintained accurate data handling using Dentrix and automated systems.