┌──────────────────────────────────────────────────────────────────────┐
│ memoona@engineering-hub:~$ cat profile.txt │
│ │
│ I build complete systems: interfaces people enjoy using, APIs that │
│ stay out of the way, data pipelines that stay clean, and models │
│ and LLM workflows that do something useful. │
│ │
│ I started in frontend, moved into Python backends, and now work │
│ across AI/ML and data engineering. Early-career, still sharpening │
│ fundamentals, and happiest when the whole stack connects. │
└──────────────────────────────────────────────────────────────────────┘
| Focus | Layer | What I do |
|---|---|---|
| Frontend | Interfaces | Responsive UIs with React, Next.js, TypeScript and Tailwind |
| Backend | Services | REST APIs with FastAPI and Node.js/Express, with authentication and RBAC |
| AI / ML | Intelligence | Scikit-learn models, RAG and agentic workflows with LangChain and LangGraph |
| Data Engineering | Foundations | SQL and NoSQL databases, data preprocessing and transformation, ingestion, ML pipelines, document indexing, import/export and reporting |
What I work with, grouped by layer. Cloud items are exposure and learning, not production mastery.
REST APIs · Authentication · bcrypt · RBAC
Machine Learning · Deep Learning (learning) · LLMs · RAG · Agentic RAG · Embeddings · Semantic Search · Prompt Engineering
SQL · Data preprocessing and transformation · Ingestion · ML pipelines · Document indexing · Import/Export · Reporting
Serverless and microservices concepts
|
Support-ticket backend Backend for managing the full ticket lifecycle, with search, filtering, attachments, import/export and reporting. Shape: client → FastAPI routes → business logic → database, with file handling and reporting around the core.
|
Document intelligence app Ask questions about your own documents. Content is indexed as embeddings, retrieved semantically, and passed to an LLM through an agent-style workflow. Shape: ingestion → embeddings → ChromaDB → retrieval → LangGraph flow → LLM → Streamlit UI.
|
|
Supervised ML project Predicts which customers are likely to leave, comparing Decision Tree and Random Forest classifiers with proper preprocessing and evaluation. Shape: raw data → preprocessing → model training → evaluation and comparison.
|
Frontend to backend Web apps connecting React/Next.js interfaces to backend services, covering authentication, role-based access and data storage. Shape: UI → API → auth and RBAC → database.
|
These are concepts I want to explore next, not completed work.
| Idea | Concept | Stack I'd reach for |
|---|---|---|
| Multi-agent research assistant | Planner, retriever and critic agents that collaborate and verify each other's answers | LangGraph, FastAPI, ChromaDB |
| RAG evaluation harness | Benchmark retrieval strategies (chunking, reranking, hybrid search) with automated quality metrics | Python, pandas, LangChain |
| Streaming data pipeline | Real-time ingestion, transformation and live analytics over an event stream | Python, PostgreSQL, Docker |
| Data observability dashboard | Track freshness, volume and schema drift across pipelines with alerts | SQL, FastAPI, React |
| AI-powered developer tool | Repository Q&A and code-review helper that understands a codebase | LLMs, embeddings, FastAPI |
| Intelligent automation hub | API-driven workflows that trigger on events and use an LLM to route and summarize | FastAPI, Node.js, serverless |
$ practice --focus fundamentals
languages : C++, Python, JavaScript, TypeScript
topics : data structures, algorithms, complexity analysis
habit : solve, review, then rewrite cleaner
Strong fundamentals make every layer of the stack easier, from query plans to retrieval pipelines.
"Make it work, make it clear, then make it connect: UI, API, data and model as one system."
| Now | Next |
|---|---|
| Agentic RAG and LangGraph workflows | Multi-agent systems and retrieval evaluation |
| Cleaner FastAPI architecture | Streaming and data pipeline design |
| ML evaluation and metrics | Deep learning foundations |
| Docker | Cloud deployment and microservices |
Open to conversations about AI/ML, full-stack and data engineering, and good learning resources.
If you're building something interesting, say hi.


