Computer Science student at Fayoum University — Faculty of Computers and Artificial Intelligence (2022–2026).
I build production-oriented AI systems: speech analytics, multimodal pipelines, RAG apps, and FastAPI backends — with attention to async design, resource limits, and clean deployment.
Path: Machine Learning → Deep Learning → NLP (DeepLearning.AI) → IBM Generative AI Engineering with LLMs (RAG, LangChain, PEFT/LoRA).
Portfolio: sober-migo.github.io
Focus areas
- Speech & multimodal: ASR, SER, Egyptian Arabic pipelines, Whisper / Wav2Vec2 / transformers
- LLMs & RAG: LangChain, Chroma, Groq, retrieval + grounded generation
- Backend / MLOps-minded: FastAPI, async jobs, Docker, GPU-aware inference
|
PDF → local ONNX embeddings → Chroma → Groq generation · Gradio Markdown UI. 
|
PACE-Egyptian_Psychiatric_Speech_Analytics FastAPI speech backend: chunking, multi-model Egyptian Arabic analysis, Docker-ready pipeline for clinical-style audio workflows. 
|
Also: HelloML · Fasten_Your_Seat_Belts (course catalog) · ML · DL · NLP · IBM GenAI
| Track | Repo |
|---|---|
| Machine Learning Specialization (Stanford / DeepLearning.AI) | Coursera_Machine_Learning_Specialization |
| Deep Learning Specialization | Coursera_Deep_Learning_Specialization |
| NLP Specialization | NLP_Specialization |
| IBM Generative AI Engineering with LLMs | IBM_Generative_AI_Engineering_with_LLMs_Specialization |
Open to internships and junior AI / ML / backend roles — speech, RAG, and applied LLMs.