San Francisco Bay Area
Results-oriented Machine Learning Lead with years of proven expertise in leading high-performing teams to deliver impactful machine learning solutions. Skilled in driving end-to-end development of cutting-edge models and algorithms, specializing in Natural Language Processing and Large Language Models. Known for fostering a collaborative team environment, providing mentorship, and leveraging technical acumen to solve complex business challenges. Passionate about staying at the forefront of emerging technologies and driving innovation in machine learning applications.
· Created a scoring service, measuring accuracy and fluency in language model outputs. Engineered with 12+ features and presented as an easy-to-use command-line script, pip library and API. Adopted across multiple Apple teams and showcased at internal summits. · Prompt engineered a large language model to produce summaries of overall sentiment and opinions in up to a million tweets and Reddit posts across different topics in a certain timeframe, incorporating batching to increase latency of response · Developed an extractive summarization tool for Apple search documents, yielding high-quality summaries and contributing to a 66% reduction in total word count. · Built and optimized a spell corrector model as a service, accurately correcting 74.4% of all messages · Fine-tuned a BERT model for NER with a focus on identifying Apple products, customer names, and locations in customer chat data. · Serve as a dedicated technical mentor, providing guidance and support to team members by sharing expertise, facilitating knowledge transfer, and offering hands-on assistance, fostering a collaborative learning environment within the team.
· Founded and led the inaugural data science consulting club on campus, establishing a platform for student engagement. · Collaborated with 3-4 companies each semester, including IBM, Copilot, and Experian, to identify suitable projects for club consultants, generating revenue of $20,000+ per semester. · Provided technical oversight for each project, guiding students in real-world project work with industry professionals. · Conducted interviews, designed data science workshops, courses and professional development sessions for Berkeley students · Currently work collaboratively with the president, staff, and students, actively contributing to club guidance and speaking at events.
Instruct 6-8 computer science students weekly through private, one-hour online lessons in Python, Java, and Scratch to teach them fundamental computer science concepts (algorithms, loops, variables, conditionals, etc.). Collaborate on team of more than 200 instructors to improve curriculum used by over 350 students.
Predicted Turn Around Time for broken Apple devices since there is currently no way to do so. Utilized regression machine learning techniques, and proper preparation of the data: selecting correct features, engineering new ones, overall improving the accuracy by 14.56 times. Constructed a deep neural network to train my data on, getting the best result of the models I selected