✅ Verified by Extern · ⏱️ In progress

BeReal Product Management Externship

Discover how product teams at social apps like BeReal design for authenticity while driving user growth. In this externship, you’ll explore what keeps users coming back— benchmarking competitors, analyzing engagement features, and designing your own growth-driven app feature. You’ll learn how to balance product innovation with BeReal’s unique mission of real, unfiltered connection. Perfect for those interested in product management, UX strategy, or social media innovation, this experience gives you a hands-on look at how product teams grow platforms without losing their core values.

Resume Intelligence System (Self-Initiated AI Product)

Owning the full product lifecycle of an AI resume-tailoring product from ideation onward: defined the problem, identified the target customer segment, wrote product requirements after validating the pain point against real market demand, and conducted market and competitive analysis (Jobscan, Teal, Resume Worded) with a pricing strategy and ROI rationale, then built a phased product roadmap that prioritizes features by customer value and delivery effort. Advancing the product through the development phase of the SDLC, managing the backlog and tracking each stage (requirements, development, QA, and testing) in JIRA to keep delivery structured and on schedule. Translating product requirements into a working MVP (Python, FastAPI, RAG with ChromaDB, Claude API) and defining a validation plan that measures real outcomes (user callbacks) over vanity metrics.

Document Intelligence Automation

Built a document-intelligence product that automated extraction and validation of key information from 200+ page mortgage documents by integrating OCR, PDF parsing (PyMuPDF), and RAG workflows into searchable, structured information. Designed the product's retrieval pipeline using LlamaIndex, applying chunk tuning, metadata filtering, and prompt optimization. Benchmarked multiple language models (Mistral, Phi-2) on accuracy and cost and packaged the result as a working demo interface and a stakeholder-facing technical report.

EV Adoption & Charging Infrastructure Optimization

Analyzed 179K+ EV registrations across 700+ Washington ZIP codes and Alternative Fuel Station data to find where charging demand outstripped supply, mapped adoption trends from 2015–2024 and built geographic pressure maps in Tableau, and translated the analysis into specific location recommendations for new charging stations.

Mental-Health Emotion & Risk Classification Using NLP and Deep Learning

Trained an end-to-end NLP pipeline on 40K mental-health Reddit posts, performing preprocessing, class balancing, and linguistic normalization; tested TF-IDF + Logistic Regression and ULMFiT AWD-LSTM, improving prediction accuracy and conducting error analysis and model diagnostics. Adopted transfer-learning workflows and fine-tuned pretrained language models on domain-specific data.

Conversational AI Project on Financial Data Analysis

Designed a conversational AI chatbot that analyzed multi-year SEC 10-K data from Microsoft, Apple, and Tesla to provide financial insights; applied data extraction and analysis techniques using pandas and Excel to compute key financial KPIs.

Malicious URL Detection

Directed a team project developing a malicious URL detection system using Convolutional Neural Networks (CNN) for feature extraction and Random Forest for classification.

Commute Time Optimization Analysis

Captured 64 real-time commute samples across two transit modes and performed time-series comparison using z-tests in R, confirming subway travel was faster than bus commutes and executing data wrangling, outlier filtering, statistical testing, and visualization.