Skip to content
View lumosityfan's full-sized avatar

Block or report lumosityfan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lumosityfan/README.md

Hi, I'm Jeff Xie 👋

Backend & Cloud Software Engineer — I build AI-powered apps and the cloud infrastructure they run on (Python, AWS, Terraform, CI/CD).

Columbia University '19 · University of Houston–Clear Lake '25

Website · LinkedIn


What I'm working on

  • Building eia-etl-pipeline, a serverless ETL pipeline on AWS for U.S. electricity demand data from the EIA API: Lambda extract and transform into an S3 bronze/silver data lake, an idempotent load into Postgres with SNS notifications, and all infrastructure in Terraform. Airflow orchestration, tests and a dashboard are next.
  • Building LLM agents and multi-agent systems with LangChain and LangGraph
  • Deploying everything as code: Terraform, GitHub Actions, serverless on AWS

Featured projects

Project What it is Stack
langchain-research-assistant-agent Research agent with 3-layer memory (short-term, FAISS long-term, episodic), retries, guardrails and an automated evaluation harness Python, LangChain, LangGraph, FAISS
simple-multiagent-system Supervisor-style multi-agent assistant that splits a request across 8 specialist agents in parallel; reachable from Discord and a Streamlit UI Python, LangGraph, FastAPI, Postgres
job-match-calculator Upload a resume, add a job description, and get a match score with strengths, gaps and suggestions Flask, OpenAI API
cloud-resume-challenge + backend My resume site on S3 + CloudFront with a serverless API, all provisioned with Terraform and deployed by CI AWS, Terraform, Lambda, DynamoDB, Playwright
eia-etl-pipeline (in progress) Batch pipeline for hourly U.S. electricity demand and day-ahead forecast data from the EIA API. Raw JSON lands in an S3 bronze layer, is transformed into curated Parquet (actual vs. forecast demand and forecast error) in a silver layer, then upserted into Postgres with an SNS notification. Terraform modules provision everything; Airflow orchestration, tests and a dashboard are next Python, AWS Lambda, S3, RDS Postgres, Secrets Manager, SNS, Terraform
page-tracker Flask + Redis service with unit/integration/e2e tests, a linted multi-stage Docker build, and a CI pipeline that publishes to Docker Hub Flask, Redis, Docker, GitHub Actions

Tech I work with

Python AWS Terraform Docker GitHub Actions FastAPI Flask LangChain Redis PostgreSQL

Get in touch

The best way to reach me is LinkedIn.

Pinned Loading

  1. langchain-research-assistant-agent langchain-research-assistant-agent Public

    Research agent built with LangChain and LangGraph: web-search tools, RAG-style long-term memory (FAISS), retries and guardrails, and an automated evaluation framework.

    Python

  2. simple-multiagent-system simple-multiagent-system Public

    Supervisor-style multi-agent assistant built with LangGraph: parallel specialist agents, retries and persistent memory, with Discord and Streamlit front ends.

    Python

  3. job-match-calculator job-match-calculator Public

    Flask app that scores how well an uploaded resume matches a pasted job description, with strengths, gaps and suggestions (OpenAI).

    JavaScript

  4. cloud-resume-challenge cloud-resume-challenge Public

    Resume website on AWS (S3, CloudFront) with Terraform IaC and CI/CD via GitHub Actions, plus Playwright end-to-end tests.

    HTML

  5. cloud-resume-challenge-backend cloud-resume-challenge-backend Public

    Serverless backend for my resume site on AWS: Lambda, API Gateway, DynamoDB and Terraform, plus LLM-powered resume tools with rate limiting

    Python

  6. eia-etl-pipeline eia-etl-pipeline Public

    Serverless AWS ETL pipeline for hourly EIA electricity demand and forecast data: Lambda, S3, Postgres and SNS, all in Terraform.

    HCL