Early engineer on the LEAP team building human-in-the-loop feedback systems for optimizing performance in our AI compliance agent, transforming complex regulations into structured LLM feedback loops that continuously improve model outputs and reduce compliance risk for our enterprise customers.

• Designed, shipped, + iterated on AI-automated workflows for legal agent deployment teams, reducing client onboarding time by 70%
• Built custom evaluation benchmarks and validation datasets with regulatory domain experts, improving agent accuracy on document processing + compliance analysis.
• Led development of microservices in Python and FastAPI, scaling to thousands of concurrent tasks under p99 latency SLAs.
• Designed PostgreSQL schemas, indexes, query patterns for document metadata + compliance audit logs, maintaining performance at scale

Developed customer-facing security tooling on Rapid7's Insight Cloud platform — visualizing high-volume threat data and shipping features safely to enterprise accounts through progressive rollouts.

• Processed + visualized high-volume security event data (100K+ events/day) in customer-facing dashboards, designing efficient query patterns + data aggregation strategies
• Implemented client-side state management for complex security dashboard views using Redux, handling real-time data updates across 10K+ concurrent user sessions
• Drove adoption of feature flagging for progressive rollouts in a security-critical environment, enabling safe deployment of new detection rules to enterprise accounts

Started career building internal tools and automating workflows. First exposure to full-stack development, API design, and production deployment.

• Analyzed Trello Power-Up usage data to prioritize which board actions to surface in Slack, focusing on the top 5 actions that covered ~80% of daily user workflows
• Owned the product spec for Trello's Slack integration, defining notification logic, card creation flows, and board action commands that shipped to 10M+ Trello users
• Built functional prototypes of Slack bot interactions (slash commands, interactive buttons, card previews) using Node.js and Slack's Block Kit API to validate UX flows before engineering handoff
• Mapped the end-to-end user journey across Trello and Slack, identifying 3 critical drop-off points in the notification-to-action flow and proposing solutions that improved estimated engagement by ~20%
• Coordinated cross-team alignment between the Trello product team and Slack partnerships team, running weekly syncs and maintaining a shared decision log to resolve 10+ integration scoping conflicts

Started career building internal tools and automating workflows. First exposure to full-stack development, API design, and production deployment.

• Instrumented product metric pipelines using Amplitude, writing custom SQL queries to track feature adoption across 15K+ users
• Built internal analytics dashboards to surface real-time KPIs, reducing stakeholder reporting overhead by 60%
• Designed A/B testing framework with engineering, specifying event schemas and statistical significance thresholds for feature rollouts
• Conducted competitive analysis of 5+ collaboration tools and synthesized findings into a product brief that informed the integration roadmap, presented to senior product leadership
• Led user research with 15+ enterprise customers to identify pain points in the documentation-to-issue-tracking workflow, surfacing insights that shaped the embed experience and inline editing capabilities

Collaborated with researchers + clinicians at Memorial Sloan Kettering to build 15+ data analysis and visualization tools for single-cell RNA sequencing workflows

• Collaborated with researchers + clinicians at Memorial Sloan Kettering to build 15+ data analysis and visualization tools for single-cell RNA sequencing workflows
• Built reproducible analysis pipelines processing high-dimensional data (20K+ genes × thousands of cells) using Pandas, NumPy, scikit-learn, and custom dimensionality reduction methods
• Designed eval frameworks validating cluster quality against known markers, establishing statistical baselines for anomaly detection

Researched novel applications of Markov Chain Monte Carlo methods to graph partitioning problems in the context of political redistricting, resulting in a co-authored publication in a peer-reviewed mathematics journal.

• Modeled redistricting as a graph partitioning problem, implementing Markov Chain Monte Carlo sampling algorithms to explore exponential combinatorial state spaces
• Built full analysis pipeline in Python (NetworkX, NumPy, Matplotlib) computing transition probabilities via spanning tree theory and analyzing stationary distributions

Student-run, revenue-generating consulting organization serving clients across industries. Worked as Associate Sales Director, Project Lead and Managing Director.

• Managed a team of student consultants across multiple concurrent client engagements, owning project scoping, delivery timelines, and stakeholder communication.
• 2x'd annual revenue by building a targeted outbound sales pipeline, closing 10 new client engagements in one year.

Offer both group and individual tutoring and mentoring in computer science, data science, and quantitative subjects to students across the 5C's.

• Mentored 20+ students one-on-one in Python, data analysis, and introductory ML — translating abstract concepts into concrete, working code.
• Designed and ran weekly workshops on foundational CS topics, emphasizing hands-on problem-solving over lecture-style instruction.