Skip to content
View anna123123123-creator's full-sized avatar

Block or report anna123123123-creator

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
Sam — Software Engineer, full-stack web and AI integration

Hi, I'm Sam 👋

Software Engineer with 5 years of fully remote experience, working with distributed teams across time zones. I own the whole path: the interface, the API behind it, the schema, and the container it ships in — most recently a retrieval-augmented search service and a client-side counterpart to it.

📈 Highlights

  • Built rag-client — the React 19 + Next.js 16 + TypeScript front-end for it: one typed API boundary, accessible states for loading, empty and error, and search-term highlighting. Pushed as its own repo with a passing production build.
  • Built rag-server — a FastAPI + SQLite retrieval service: sentence-aware chunking, server-side ONNX embeddings, a REST search API, cascading deletes, a Dockerfile that bakes the model into the image, and a measured confidence threshold so the API can answer "nothing here" instead of guessing.
  • Built Browser RAG — retrieval-augmented search that runs entirely in the client: own chunking with sentence overlap, local MiniLM embeddings on WebAssembly, cosine ranking. No server, no API key, nothing uploaded.
  • Integrated LLM-powered conversational assistants and automated content pipelines into client products, writing the retrieval and context layers myself.
  • Shipped 50+ open-source browser tools and publish a catalogue of 60+ business-system and AI source-code products through my own storefront, handling delivery and after-sales directly.
  • 5 years of remote contract work with international clients, coordinating asynchronously across time zones and documenting decisions in writing.

🧰 Toolbox

  • Back-end & data — Python, FastAPI, REST API design, SQLite (schema design, foreign keys, indexes, cascading deletes), Docker and docker-compose, ONNX Runtime
  • AI integration — retrieval-augmented search (chunking, embeddings, cosine ranking), LLM API integration, prompt and context design, measured confidence thresholds
  • Front-end — JavaScript (ES6+), HTML/CSS, Canvas and SVG rendering, Web Audio and Web Speech APIs, responsive UI, no-build-step delivery
  • Design & content — Figma to production code, SEO, web editing
  • Also work with — Node.js, PostgreSQL, React, Next.js, TypeScript, Git

🧪 Built in the open

Browser tools I ship in public — no backend, no signup, no build step, each with a live demo:

Project Where
rag-server — retrieval as a service: FastAPI + SQLite, server-side embeddings, REST API, Dockerised source
rag-client — the React 19 + Next.js 16 + TypeScript client for that API source
Browser RAG — asks your own documents a question and returns the passages that answer it, entirely client-side demo
FAQ bot — keyword-matching Q&A assistant open
Outline → mind map — parses an indented outline into a tree and lays it out in ~200 lines of vanilla JS open
Outline → slides — outline to slide deck preview open
Resume ⇄ JD matcher — keyword match scoring open
Text to speech — reads text aloud in the browser open

54 public repositories, 50 of them with a live demo.


📫 Get in touch

Open to full-time remote roles and contract work: LLM integration, Next.js builds, and Figma-to-code.

Pinned Loading

  1. resume-match resume-match Public

    Browser-only resume-to-job-description keyword matcher: scores the match and lists the missing keywords. No server, no signup.

    JavaScript

  2. faq-bot faq-bot Public

    Keyword-matching Q&A bot that runs entirely in the browser. Configure a question/answer table and it works - no AI API required.

    JavaScript

  3. outline2mindmap outline2mindmap Public

    Turn an indented text outline into a visual mind map in the browser. Exports PNG/SVG. No install, no signup, no backend.

    JavaScript

  4. outline2slides outline2slides Public

    Turn a text outline into a slide preview, and export Marp Markdown that converts straight to PPTX. Runs entirely in the browser.

    JavaScript

  5. browser-rag browser-rag Public

    Retrieval-augmented search that runs entirely in the browser: chunking, local embeddings and cosine ranking. No server, no API key.

    JavaScript

  6. rag-server rag-server Public

    FastAPI + SQLite retrieval service: ingest documents, embed them server-side with ONNX MiniLM, expose a REST search API. Dockerised, tested, measured confidence threshold.

    Python