Building
agentic systems
that ship.
I'm Mohammed Ajzel — an AI/ML Engineer and Data Scientist turning complex data into production systems. I build Agentic AI, RAG pipelines, fine-tuned LLMs, and analytics dashboards that drive measurable outcomes.
From data to intelligence — across AI, analytics, and the full stack.
Agentic AI Systems
Multi-agent architectures with LangGraph, LangChain, and the Model Context Protocol. ReAct reasoning across tool servers, deployed as FastAPI services.
Retrieval-Augmented Generation
Production RAG pipelines with FAISS vector search, dual embeddings, and sub-1.5s query latency across multi-document corpora.
LLM Fine-Tuning
LoRA / QLoRA fine-tuning with Unsloth on Phi-3-mini. 4-bit quantized training, GGUF export, offline inference via Ollama.
ML & Data Science
Statistical modeling, EDA on 10k+ records, A/B testing, and classification pipelines with measurable business outcomes. SQL, Pandas, NumPy, Scikit-learn.
Data Analytics & Visualization
End-to-end analytics workflows: data cleaning, feature engineering, dashboarding, and storytelling with Tableau and Power BI. Translate raw data into board-ready insights.
AI systems & data pipelines.
Real users. Real numbers.
MCP LangChain Agentic System
LLM Fine-Tuning Pipeline
AI Document Searchbot
Cloud-Native ATS Analyzer
Data Science & Analytics Portfolio
Deepfake Image Detection
AI News Live
Industry experience, AI evaluation, and applied analytics.
AI Evaluation & Benchmarking Contributor
Designed and validated multimodal LLM benchmark tasks spanning visual reasoning, scientific figures, and quantitative inference. Authored image-grounded prompts, golden solutions, and evaluation criteria; analyzed model failures and refined workflows with automated quality checks.
AI Automation Extern
Engineered an n8n pipeline using Claude and Gemini to collect and standardize product data across 48+ SKUs, reducing manual collection by 70%. Automated competitor analysis and delivered a 19-page pricing and strategy report in under two hours, with summaries organized through Google Drive.
AI Engineer Intern
Built an NLP-driven enterprise email categorization and prioritization system using transformer models and prompt engineering. Delivered a production-ready classification workflow integrating LLMs and enterprise AI design; certified by the Infosys SVP and Head of Education, Training and Assessment.
Machine Learning / Data Analyst Intern
Benchmarked three models on 10,000+ records for cuisine classification; Random Forest reached 90%+ accuracy, 12% above baseline. Deployed a Flask recommendation API returning personalized Top-20 results across cuisine, rating, city, and price filters.
AI, ML & Data Science Intern
Deployed a Flask energy prediction API with a Random Forest model (92% R²), serving 100+ predictions per second. Built a real-time Chart.js dashboard and automated retraining on uploaded datasets, cutting manual analysis by 40% and improving forecast accuracy by 30%.
BCG GenAI Job Simulation
Built an AI-powered financial chatbot and analyzed 50+ data points from 10-K and 10-Q reports with Python and Pandas to deliver structured investment insights.
Tata Group Data Analytics Simulation
Performed exploratory analysis on 10,000+ records using GenAI tools and identified risk indicators with 90% accuracy.
Lloyds Banking Group Data Science Simulation
Built and tuned a Random Forest churn model (ROC-AUC 0.82), using GridSearchCV and feature importance to identify key churn drivers.