software engineering / applied AI / backend systems / simulations
I am a Colby College computer science student preparing for Summer 2027 software engineering internships. I like building products where the behavior can be tested, measured, explained, and made useful for real people.
| Field | Value |
|---|---|
| Looking for | Summer 2027 SWE internships |
| Strong fits | backend infrastructure, applied AI products, developer tools, fintech systems, full-stack products |
| Recent work | LinkedIn SWE Intern, Wayfair SWE Intern, Colby ML Research Assistant |
| Academic base | Colby College, CS major, Math minor, expected May 2028 |
| Working style | ship useful systems, test behavior, write clear docs, explain model decisions |
| Project | What to look for | Stack |
|---|---|---|
| LOBster | Limit order book engine with price-time matching, O(1) cancellation, market microstructure analytics, backtesting, tests, and CI. | Python, pytest, NumPy, SciPy |
| ClearMarket | Full-stack market intelligence dashboard with stock data, news sync, and AI-assisted market context. | TypeScript, React, Express, Supabase |
| 0-KM | Team-built mobile app for helping long-distance couples stay connected. | TypeScript, React Native, Expo, Node.js |
| Traffic Sign Classifier | PyTorch CNN and Streamlit app for classifying German traffic signs across 43 classes. | Python, PyTorch, Streamlit |
| Molecule Semantic Search | Natural-language and SMILES-based molecular search with embeddings and graph exploration. | React, embeddings, cheminformatics |
2026 / LinkedIn
LLM agent for audience targeting and customized A/B testing workflows.
Built DSL compilation, catalog-backed semantic validation, and LangGraph orchestration.
2025 / Wayfair
Real-time retail analytics dashboard deployed across 50 global stores.
Refactored GraphQL mutations and reduced search lookup time by 26%.
2025 / Colby College
Neural network loan prediction with decision-tree surrogate explanations.
Improved human prediction accuracy by 32%.
2025 / IFMBE Proceedings
Comparative study of BERT-family transfer learning for early Alzheimer's diagnosis.| Area | Tools |
|---|---|
| Languages | Python, Java, Kotlin, JavaScript, TypeScript, SQL |
| AI / data | LangGraph, PyTorch, SHAP, embeddings, RAG |
| Web / backend | React, Next.js, Node.js, FastAPI, gRPC |
| Infra | AWS, Docker, PostgreSQL, MongoDB, GitHub Actions |
Clear interfaces. Measurable behavior. Systems that recover gracefully. AI features that are evaluated instead of hand-waved. Documentation that helps the next engineer move faster.