Skip to content
View mukeremshifa's full-sized avatar
0 followers · 3 following

Achievements

Achievement: Pull SharkAchievement: YOLO

Achievements

Achievement: Pull SharkAchievement: YOLO

Highlights

Block or report mukeremshifa

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.

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
mukeremshifa/README.md

ASCII portrait of Mukerem Shifa beside a terminal reading: whoami, Mukerem Shifa, full-stack developer, AI applications. Building ConverseKit and SynapseDeck.

About: I like the problems that hide behind the demo. Anyone can call an LLM API. The work is everything around it: deciding which passage is worth retrieving, keeping one client's documents out of another client's answers, and holding a conversation together when the stream drops halfway through a sentence. I am early in my career and I learn by shipping whole systems rather than tutorials. The four projects below are deployed, tested and documented. Open to junior and entry-level roles.

Selected work: four projects, all deployed or tested

ConverseKit, live. Multi-tenant AI chat that installs with one script tag. Answers from each client's own documents, eleven LLM vendors behind one interface, tenants isolated by row-level security rather than by application code. TypeScript, Cloudflare Workers, Supabase, pgvector, Hono.

SynapseDeck, live. Notes in, flashcards out, reviewed on a real FSRS scheduler. Cards stream in as the model writes them and pass a review gate before entering a deck. Every figure on the progress page is counted from an append-only log. React 19, TypeScript, Supabase, Edge Functions, ts-fsrs.

Document RAG QA Bot, capstone. Question answering grounded in uploaded PDFs: chunking, embedding, vector retrieval and generation that cites what it read. Built as the capstone for IBM's AI engineering coursework. Python, LangChain, Gemini, ChromaDB.

Little Lemon API, capstone. Restaurant back end covering all 21 acceptance criteria of the Meta capstone: role-based permissions across four user groups, cart and order flows, throttling, and 24 acceptance tests. Python, Django, Django REST Framework.

Stack. Languages: TypeScript, Python, JavaScript, SQL, Java. Frontend: React 19, Vite, Tailwind, TanStack Query, Zod. Backend: Node, Hono, Django REST, Cloudflare Workers. Data: PostgreSQL, Supabase, pgvector, row-level security. AI: LangChain, OpenAI, Anthropic, Gemini, Groq, RAG. Tooling: Vitest, GitHub Actions, Git, Linux, Docker.

Languages across public repositories, 1,661,208 bytes total: TypeScript 57.9 percent, JavaScript 26.5 percent, Python 4.9 percent, PLpgSQL 4.1 percent, HTML 3.4 percent, CSS 1.2 percent, Java 1.1 percent, SCSS 0.9 percent. Trajectory by repository creation year: 2024 all Java, 2025 mostly TypeScript, 2026 TypeScript with Python appearing.

Contact

Email mukeemoha@gmail.com LinkedIn, mukeremshifa mukeremshifa.com GitHub, mukeremshifa

Pinned Loading

  1. conversekit conversekit Public

    Drop-in AI chat widget for any website. Multi-tenant, eleven AI vendors behind one interface, RAG over your own docs. Cloudflare Workers + Supabase.

    TypeScript 2

  2. ibm-capstone-rag-bot ibm-capstone-rag-bot Public

    A Document-Based RAG QA Bot utilizing LangChain, Google Gemini, and Chroma DB

    Python 1

  3. synapse-deck synapse-deck Public

    Paste your notes, get flashcards, and review them on a real FSRS spaced-repetition scheduler. Cards stream in as the model writes them and pass a review gate before entering a deck. React 19 + Type…

    TypeScript 1

  4. gamified-servey-project gamified-servey-project Public

    Survey Quest, a React prototype that turns a questionnaire into a game with XP, levels, badges and a confetti finish. XP is awarded for participation only, never for which option you pick, so the g…

    JavaScript 1