LLM Terraform Code Generation: Built an AI-driven system for generating Terraform code blocks using AWS Bedrock and OpenSearch. Designed document crawlers using Crawl4AI, implemented batched Lambda pipelines, and configured a Terraform MCP server for the project’s infrastructure.

Graph-Driven Workflows: Currently developing a feature leveraging AWS Neptune to enable task tracking, user assignment, and entity relationships; responsible for node/edge upset logic and CRUD integration into AI repository.

AWS Neptune Graph Database: Designed architecture and implemented Neptune Connection Class and startup scripts using Python to introduce graph database support for AI and relationship-driven features. Established permission models, CRUD safety guards, and multi-environment separation to ensure data isolation and system integrity.

System Security Plan (SSP): Independently led end-to-end development of a System Security Plan module aligned to CMMC compliance frameworks, spanning infrastructure (Terraform), backend (API layer), and frontend. Created a critical feature that directly supported acquisition of federal contracts by enabling customers to certify compliance.

AI Assistant: Created an AI-powered assistant to handle evidence attachment queries using cosine similarity for response accuracy between query and answer embeddings. Delivered a tool with an 80% improvement in query response accuracy for evidence requests.

Asynchronous Embedding System: Architected and implemented a decoupled system for embeddings that reduced processing time by over 90% providing a scalable workflow. Developed and maintained

asynchronous ML inference pipelines for embedding generation using AWS SageMaker, Bedrock, and

PGVector, enabling scalable semantic search capabilities across millions of data points.

Automated Evidence Testing and Document Comparison: Developed features for summarizing and highlighting differences in evidence using LLMs, cutting manual review time by 70%.

Canonical Database & API Development: Designed and implemented a canonical database to serve as the single source of truth for customer resource versions. Built APIs to enhance data consistency, reducing manual input by 50%.

Terraform Integration Expansion: Spearheaded full-stack project, adding Gitlab/Github integrations, improving integration options by 200% and completed 10 weeks early.

Customer Interaction Tracking: Developed an append-only logging service, now logging

thousands of interactions, increasing tracking visibility over 98% through added Mixpanel analytics

Ada’s mission is to prepare women and gender expansive adults to be software developers while advocating for inclusive and equitable work environments. We prioritize serving Black, Latine, Indigenous Americans, Native Hawaiian & Pacific Islander, LGBTQIA+, and low-income people.

Our cost-free training program includes a collaborative learning environment, individualized support (mentors, tutors, a network of mental health supports, affinity groups), and an applied learning internship.