Skip to content
View suhasaitham22's full-sized avatar
2 followers · 0 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Block or report suhasaitham22

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. Markdown is supported. 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
suhasaitham22/README.md

Hey, I'm Suhas 👋

Software Engineer (GenAI) @ Amazon · Seattle, WA

I work on production AI systems: LLM orchestration, RAG pipelines, and agentic workflows. Before Amazon I was a data engineer at Humana and a data scientist at Equifax. I did my masters in Data Science and Business Analytics at UNC Charlotte, where I also picked up two research publications along the way.

What I've built

ApplyBloom

My main side project. A full-stack auto-apply platform: Next.js frontend, a stateless Cloudflare Worker API, and async job processing pipelines. It discovers jobs and ranks them against your resume using vector embeddings and RAG, runs on cron, and ships with a Chrome (MV3) extension and a Supabase backend.

DxChat · live demo

An AI chatbot that answers general questions about health symptoms, built with Streamlit and open-source LLMs through the Hugging Face Inference API. It crossed 500 interactions in its first week. Strict guardrails and clear disclaimers throughout: general info only, never a diagnosis.

Offer Similarity Analysis

A Streamlit app that finds and compares similar retail offers across brands, categories, and retailers. Uses NLP text similarity (cosine similarity) plus data merging to match offers across messy retailer datasets.

RAG-based-Chatbot

A Python chatbot that lets you chat with your own documents using embeddings and an LLM. Classic RAG, built to learn the moving parts.

Tech I work with

Languages & frameworks Python TypeScript Java React Next.js SQL PySpark

AI & ML Hugging Face PyTorch scikit-learn pandas Jupyter Streamlit

Cloud & data AWS Google Cloud Azure Docker Snowflake Databricks PostgreSQL MongoDB Redis

GitHub stats

GitHub streak

followers stars profile views

Publications

  • Human-Computer Interaction, International Journal of Engineering Applied Sciences and Technology (IJEAST)
  • Artificial Intelligence in Radiology: The Growth of Ecosystem Opportunities and Challenges, Design Engineering

Right now

  • Building GenAI systems at Amazon (LLM orchestration, RAG, agentic workflows)
  • Growing ApplyBloom on the side
  • Always going deeper on LLM system design and evaluation

Say hi

Pinned Loading

  1. ApplyBloom ApplyBloom Public

    AI-assisted job application platform: Next.js frontend, Cloudflare Worker API, async apply pipelines, LLM/RAG job matching.

    TypeScript 3 1

  2. DxChat DxChat Public

    AI chatbot for general symptom information, built with Streamlit and open-source LLMs via Hugging Face. Informational only.

    Python

  3. heart-stroke-prediction heart-stroke-prediction Public

    Predicting heart disease and stroke risk with machine learning (Jupyter notebooks).

    Jupyter Notebook

  4. mall-customer-segmentation mall-customer-segmentation Public

    Customer segmentation on the Mall Customers dataset using clustering.

    Jupyter Notebook

  5. offer-similarity-analysis offer-similarity-analysis Public

    Streamlit app that finds similar retail offers using NLP text similarity and data merging.

    Python 1

  6. RAG-based-Chatbot RAG-based-Chatbot Public

    Chat with your own documents using retrieval-augmented generation (embeddings + LLM).

    Python 1