Skip to content
View adilzubair's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report adilzubair

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
adilzubair/README.md

Hi, I'm Muhamed Adil

AI Engineer at CyberCore Technology · Abu Dhabi, UAE · muhamedadil.com

I build LLM systems that hold up in production — retrieval, agents and fine-tuned models — and run the GPU infrastructure that serves them.

Python Go FastAPI PyTorch LangChain Ollama Kubernetes Docker PostgreSQL React

Now

  • Building internal AI applications at Cyber Core, from implementation through deployment
  • Running a GPU-backed Kubernetes cluster: bringing up GPU nodes, and deploying and serving models on it
  • Previously: LLM decision support, evaluation and LoRA fine-tuning at Greentruth; Go/Python services at Cooee; OpenVINO inference at Intel

Featured

Project What it is
Cortex Multi-agent coding assistant for your own codebase — LangGraph orchestration, incremental SHA-256 indexing, LSP analysis and reranked retrieval. Runs on Ollama or OpenAI.
Loomin-Docs Air-gapped collaborative editor with a grounded AI assistant: FAISS retrieval over uploaded files, PII masking, Ollama models, and an offline Docker bundle for RHEL 9.
Scout · site Private launcher for Windows 11 with optional on-device AI — decision models for ranking and intent, EmbeddingGemma for multimodal search. Public beta.
FindIt · site Offline multimodal search that travels with an external drive: on-device embeddings for photos, video, documents and audio, Whisper transcription, face search, and hybrid ANN + full-text retrieval.
AI Harness Basics An agent harness built up in five runnable stages — tool loop, guardrails, environment, verification and retries, deterministic recovery.

Also built

  • agents-from-scratch — agent patterns from the ground up: tool use, RAG, routing, parallelization, orchestration
  • local-rag-agents — local RAG systems with Docling, LangGraph, FastAPI and ChromaDB
  • phi3-medical-qlora — QLoRA fine-tuning of Phi-3-mini on medical Q&A, plus GPT-2 LoRA and DistilBERT PEFT experiments
  • Bitmasters_Intel_LLM — TinyLlama optimised with OpenVINO for CPU inference on Intel AI laptops
  • Odin — multi-agent trading companion with autonomous trading and alerts
  • deckspring · site — create and edit slide decks with AI, hosted or local, with editable PowerPoint export
  • iftarinuae · site — community platform for Iftar spots across the UAE: 150k+ views, 40k+ visitors, close to 200 places added by the community

Research & writing

Activity

GitHub contribution graph

Connect

Website LinkedIn X Email

Pinned Loading

  1. cortex cortex Public

    Cortex is a sophisticated AI coding assistant that combines the power of multi-agent orchestration with Retrieval-Augmented Generation (RAG) to provide intelligent, context-aware answers about your…

    Python 10 4

  2. agents-from-scratch agents-from-scratch Public

    A comprehensive collection of patterns, building blocks, and workflows for building AI agents from the ground up using Python. This repository serves as a practical guide to understanding and imple…

    Python 1

  3. Yukthi_hackathon Yukthi_hackathon Public

    This repository contains the project built during the Yukthi'24 Hackathon

    JavaScript 1 1

  4. QuickEval QuickEval Public

    Forked from Project-SCE/QuickEval

    This repository contains the website made as part of mini project.

    JavaScript

  5. Bitmasters_Intel_LLM Bitmasters_Intel_LLM Public

    This repository contains the files of the project titled "Running GenAI on Intel AI Laptops and Simple LLM Inference on CPU and fine-tuning of LLM Models using Intel® OpenVINO"

    Python

  6. Crashchain Crashchain Public

    Forked from Project-SCE/Crashchain

    A blockchain-integrated vehicular forensic investigation tool designed to capture and verify crash data securely. The system includes features like accident reports, real-time OBD data, and crash f…

    JavaScript