Skip to content
View Sami-7264's full-sized avatar

Block or report Sami-7264

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
Sami-7264/README.md

Sami Ullah Butt | AI Engineer

I build AI agents, voice AI, and LLM-powered features inside production SaaS products. Six years in full-stack and backend engineering, the last three focused on AI and LLM systems. I work end to end: architecture, backend, frontend, and deployment.

What I build

  • AI agents and tool use: agents that act on real business systems, including MCP servers
  • Voice AI: realtime spoken agents on LiveKit and the OpenAI Realtime API, plus speech-to-text pipelines with Whisper
  • LLM integrations: generation, scoring, structured extraction, summarization, and document OCR
  • Multi-tenant SaaS: tenant isolation, RBAC, subscriptions and billing, admin tooling

Tech stack

  • Languages: TypeScript, Python, PHP, JavaScript
  • Backend: Node.js, Fastify, Express, AdonisJS, Laravel, FastAPI
  • Frontend: Next.js, React, Vue, Tailwind CSS
  • AI: OpenAI API (Realtime, Responses, vision, TTS, Whisper), Anthropic API, LiveKit Agents, Model Context Protocol, Faster Whisper, Gemini
  • Data and infrastructure: PostgreSQL, MySQL, Redis, BullMQ, Prisma, Drizzle, Docker, Stripe

Open-source projects

Project What it does Stack
voice-commerce Voice-command e-commerce SaaS: manage a store, launch ads, and hear analytics through spoken commands Next.js, Express, Prisma, FastAPI, Faster Whisper, Anthropic
skool-notes Chrome extension and API that capture lesson transcripts and generate AI summaries TypeScript, Fastify, Drizzle, PostgreSQL
file-to-md Converts documents, images, audio, and video links to Markdown FastAPI, React, markitdown, Tesseract, Whisper
video-transcriber Local transcription toolkit with txt, srt, vtt, and json export Python, Faster Whisper, FFmpeg
rss-sports-feed RSS sports ticker that renders broadcast-style PNG frames for LED media players TypeScript, Node.js, Vercel

Case studies

Private projects, described at the architecture level.

More private work

  • QuickBooks MCP server: connects Claude to QuickBooks Online through 13 accounting tools with OAuth
  • AI proposal assistant SaaS: job analysis, proposal generation and scoring, reply drafting, and a browser extension
  • HR management system: attendance, leave, payroll, and performance on Fastify, Drizzle, and PostgreSQL, with automated tests and CI
  • Job board scraper: BullMQ queues, Playwright fallback, and an admin dashboard

Experience highlights

  • 6+ years building web products across backend, frontend, and mobile
  • 3+ years shipping AI and LLM features in production products
  • Built multi-tenant SaaS platforms with tenant isolation, RBAC, and subscription billing
  • Shipped a realtime voice AI agent that teaches live classes

Contact

Open to AI engineering roles and freelance projects.

Pinned Loading

  1. file-to-md file-to-md Public

    Convert documents, images, audio, and video links to Markdown with a FastAPI backend and React UI, built on Microsoft markitdown

    Python

  2. rss-sports-feed rss-sports-feed Public

    TypeScript

  3. skool-notes skool-notes Public

    Chrome extension and API that capture course lesson transcripts and generate AI summaries, stored per lesson in Postgres

    TypeScript

  4. video-transcriber video-transcriber Public

    Local media toolkit: Faster Whisper transcription to txt, srt, vtt, and json, plus GIF frame extraction and contact sheets

    Python

  5. voice-commerce voice-commerce Public

    Voice-command e-commerce SaaS: manage a Shopify store, launch Facebook ads, and hear analytics through natural-language voice commands

    TypeScript

  6. local-pdf-converter local-pdf-converter Public

    Offline PDF converter: PDF to Word (DOCX), TXT, PNG/JPG, and images to PDF. Python CLI with optional GUI.

    Python