Architected a production-grade agentic AI system featuring fine-tuned models, LLM-as-judges, task evaluators, safety guardrails, and observability, expanding the platform toward universal coverage.

• Increased coverage 20×, enabling a clear path toward 4,000+ future integrations

• Developed fine-tuned OpenAI models with LangFuse traces to generate high-quality training data

• Curated evaluation and golden datasets and built Kiln-based task-performance evaluators to benchmark fine-tuned vs. frontier models

• Used LangFuse trace-level observability to support logging, debugging, and iterative refinement

• Introduced safety guardrails, including risk detection heuristics, to ensure the agent operated within defined boundaries.

* Designed prompt-engineered LLM-as-judge to score task outcomes per risk, generate confusion matrices, and systematically eliminate false negatives.

• Redesigned the integration architecture to eliminate ~90% of per-integration code, dramatically simplifying large-scale onboarding

• Led design and implementation of Rightfoot Public API v1.0.0 delivering 83% ARR growth and 50% customer expansion, including securing the company's largest deal.

• Drove $1.5M projected annual savings by self-initiated optimizations: 60% faster balance processing, combining API operations, and eliminating third-party services.

• Architected an extensible transaction processing system supporting major financial institutions including Chase (22% of total transaction volume), enabling the company's first pre-PMF customer acquisition and establishing a new B2B revenue stream.

• Won company-wide 2024 Hackathon with BalanceIQ, a full-stack PoC for predictive balance calculations to reduce NSF penalties driving 2024 product roadmap.

• Scaled system throughput 2X by optimizing tokenizer chunking with 4x scale potential.

• Architected batch completion through Stream Aggregators to resolve concurrency issues completing thousands of batches.

• Automated batch results with a Retool 1-click CSV tool (1hr → 5s) and Slack-webhook notifications saving engineering hours and improving cross-team communication.

• Prevented 17% data loss for unsupported institutions, resolving 31% of batch timeouts, reducing batch completion SLA from 60 to 5 minutes for affected batches

• Enabled loan underwriting use cases by delivering Balance Errors v0 and v1.

• Mentored engineers on ROI-focused technical decision-making, elevated quality through rigorous code/design reviews, CI/CD linting, extensive refactoring and modularization, resolved API inquiries in <24 hours boosting customer NPS by 40%

• Led a team of nine to build full-stack features in React and Rails for an experience relationship management platform.

• Conducted 1:1s and performance reviews; directed multiple promotions; fostered team unity through organized events

• Estimated and communicated timelines of multi-quarter projects to stakeholders in collaboration with lead engineers.

• Reviewed technical specs, conducted code reviews; fixed several bugs to not detract focus of the team.

• Collaborated with product, design and engineering leadership on quarterly roadmap planning and bi-weekly sprint planning

• Led evaluation and selection of a translation system to streamline developer/QA processes for new language adoption.

• Led and developed several technical initiatives including creating a CircleCi workflow to run Cypress end-to-end tests gating production releases and using Datadog Synthetics providing Slack uptime alerts for production/QA environments.

• Coded full-stack features including a language dropdown translating an experience list into a chosen language and customized and specific error messaging for experience searches.

• Extensively modernized repositories: added testing libraries, HTML parsers, commonized styles, linters and pre-commit hooks. Removed dead, deprecated code and libraries, centralized FullStory snippets, and upgraded to M1-compatible gems.

• Led, planned and developed full-stack revenue-generating projects in React, Python, iOS and Android.

• Architected high performing lead generation ads and previous A/B test iteration leading to expansion of customer base beyond real estate to all SMB and improved retention

• Other features include Projects postbox for high quality DIY content, free/premium plans, and Opcity partnership

• Contributor to several full-stack/mobile features including free trials, integrated post flow, home owner dashboard, Salesforce/datalake integration, vetting libraries such as Victory to graph ad metrics, and Listhub partnership.

• Developed Android features like newsfeed ads, mobile specific designs and bug fixes.

• Contributed to the development and success of a positive and productive team culture:

• Participated in on-call and troubleshooted critical production issues.

• Onboarded and mentored engineers, performed code reviews, conducted technical interviews

• Leader and contributor to women’s and LGBT ERGs