About Me

My name is Halia, and I am a full stack developer from Maui, Hawaii. When I'm not coding, I love to spend time with family and friends or out in nature. My favorite activity is exploring the beautiful island I call home, finding secret spots to hike and new waterfalls to climb. I'm always up for navigating the unexpected— whether it's swatting bugs on the trail or squashing them in my code.

  • Image 1
  • Image 2
  • Image 3
  • Image 4
  • Image 5
  • Image 6
  • Image 7
  • Image 8
  • Image 9
  • Image 10
  • Image 11
  • Image 12
  • Image 13
  • Image 14
  • Image 15

Technical Skills

Full Stack Development

I love working across the stack, building beautiful and modern client side applications with React, Typescript, and various UI libraries (my favorite being shadcn/ui). I also focus on performance and scalability using Node.js, Typescript, and Python to build out API layers, persisting data to DynamoDB and PostgreSQL databases.

Tech stack:
TypeScript
JavaScript
Python
Node.js
Express.js
React.js
Next.js
Redux
GraphQL
HTML5
CSS
SASS
Shadcn
MUI
Aceternity UI
PostgreSQL
MySQL
DynamoDB
MongoDB
Jest

DevOps & Infrastructure

Once the application is built, I am capable to deploy it anywhere (build once, deploy everywhere). I have experience using AWS CDK v2 to configure highly scalable and secure infrastructure while making sure it's cost effective for the team.

Tech stack:
AWS CDK
aws-sdk-v3
AWS EC2
AWS Lambda
AWS ECS Fargate
AWS EKS
AWS ECR
AWS S3
AWS WAF
AWS IAM
AWS IoT (pub/sub)
AWS CloudWatch
AWS CodePipeline
Git
GitHub Actions
Docker
Kubernetes
Prometheus

GenAI & LLM

ML integration is a growing field and I'm staying at the forefront of these platforms. I have gotten a Generative AI application deployed to Prod with live customers, ensuring reliability and security each step of the way.

Tech stack:
AWS Bedrock
OpenAI
Anthropic (Claude 3, 3.5)
Cohere
Meta LLama
Foundation Models
Embeddings Models
LangChain
pgvector
Pinecone

Experience

RevApps

Remote, USA

Software Engineer

Mar 2024 - Present

  • Engineered an ML driven text analysis system using AWS Bedrock, integrating a Foundation Model (FM) with AWS Lambda and DynamoDB to process and analyze data in a serverless architecture, delivering real-time sentiment analysis for executive teams
  • Spearheaded the initial POC to create a RAG solution for specific knowledge based research, creating AWS Lambda functions with Python and LangChain, orchestrating text splitting from documents to upsert vector data into PostgreSQL with pgvector
  • Created the localized and globally Distributed Presence Awareness Engine, leveraging MQTT5 with LWT strategies to monitor multiple User Presence Types in real time on top of AWS IoT Core with WSS, scaling to millions of concurrent DAUs elastically
  • Created the ECS Fargate Autoscaling Group with AWS CDK v2, using Typescript to configure the Infrastructure as Code, measuring against CPU utilization for scale out/ scale in functionality with configurability across multiple deployment environments
  • Leveraged AWS EventBridge to orchestrate the Email Management Platform, connecting to ECS Fargate for ingesting user metadata and triggering AWS Lambda to connect to SES with Typescript on Node.js, improving the account management to reduce churn by 18%
  • Architected the Customer Engagement Microservice from scratch, leveraging AWS CDK v2 to construct an entire serverless API layer with API Gateway, AWS Lambda with Python and leveraging the boto3 library to persist interaction data in DynamoDB
  • Achieved over 92% testing coverage with our core Jest Typescript testing suite, integrating CI/CD utilizing AWS Codepipeline
  • Improved user conversion and product growth by revamping the Next.js v13 product page, utilizing the shadcn/ui and Aceternity UI libraries, in conjunction with leveraging SSR for much more performant page load times and responsiveness
  • Improved the overall application security for our core Live Updates Platform by protecting against CSRF with internally managed secure tokens, leveraging AWS Secrets Manager, and removed PII from all logs across our ECS and Lambda Cloudwatch logs