Mustafa Mirza

Worked on a Log Reasoning Agent dealing with root cause analysis on large log files

TypeScript, Python, LLM API’s (Open AI, Google Gemini, Anthropic), Docker, PostgreSQL, Node.js

15HR Reader for CS 188 Teaching discussion sections and running office hours, covering machine learning techniques like Reinforcement learning, Q-learning, Markov decision processes, Hidden Markov models, Neural Networks, Bayesian networks, and Inference techniques. Assisting students with homework, discussions, and projects. Guiding students through 5 projects focusing on building AI agents using Python and PyTorch. In charge of assisting with project 5 (machine learning) using PyTorch and NumPy

Participated and completed two 8 week intensive training programs for iOS and React Native app development. Completed 4 mini projects for both respective training programs which included using UIKit, React Native, Firebase, Typescript. Integrated API’s and produced design mockups and prototypes to strengthen designs, enhance user experiences and app interactions for a start-up.

Working on RAG teaching assistant agent - ROAR AI at UC Berkeley's Center for Augmented, Virtual, and Extended Reality research

Researched and implemented innovative data scraping techniques to extract comprehensive insights into accepted insurance plans across American hospitals. Revamped the filtering functionality, leading to an exceptional 250% reduction in the time required to locate patients' specific insurance plans. Transformed the website into an intuitive iOS app using UIKit, enhancing user engagement and accessibility while streamlining navigation.

Support weekly lab sections of 30 students to help students debug code and strengthen fundamental understanding of data structures and algorithms for lab and project assignments. Data structures covered in lab include trees, HashMaps, queues, stacks and graphs, along with runtime & memory analysis.

Assisted in the frontend development of the INTED browser extension which manipulates video speed to cater to non-native english speakers. Initiated and guided strategic discussions, orchestrating the creation of an engaging user onboarding journey. Leveraged interactive paragraphs and audio components to infuse gamification into the registration process, resulting in an increase in the conversion rate from site visitors to sign-ups. Applied data-driven strategies to elevate the precision of Cognitive Quotient (CQ) score assignment.

Relevant Coursework: • Data Structures (CS 61B) • Discrete Mathematics and Probability Theory (CS 70) • The Structure and Interpretation of Computer Programs (CS 61A) • Principles and Techniques of Data Science (DATA 100) • Designing Information Devices and Systems I (EECS 16A) • Linear Algebra and Differential Equations (Math 54) • Multivariable Calculus (Math 53) • Probability for Data Science (Data 140) • Linear Algebra (Math 110) • Efficient Algorithms and Intractable Algorithms (CS 170) • Artificial Intelligence (CS 188) • Computer Vision (CS 180) • Introduction to the Internet and Network Protocols (CS 168) • Designing Deep Neural Networks (CS 182) • Introduction to the great ideas of Computer Architecture (CS 61C) • Machine Learning (CS 189) • Operating Systems (CS 162)