Mohamed
Fathir
AI Engineer & Data Scientist
Building scalable
Automation.
Specializing in MLOps, LLMs, and Multi-Agent Systems to drive enterprise efficiency.
Visual
Intelligence.
Advanced Intelligent Document Processing and real-time computer vision systems.
About Me
AI Engineer with 3+ years of experience specializing in MLOps, Generative AI, Multi-Agent Systems, and Intelligent Document Processing (IDP).
An IT graduate from the University of Colombo, I combine a strong foundation in Industrial Statistics with practical expertise in architecting scalable AI solutions.
My work focuses on building and deploying high-throughput production systems and orchestrating autonomous workflows to drive enterprise automation. I thrive in fast-paced environments, bringing a start-up mindset and a commitment to agile delivery.

My Journey
My professional experience and academic background.

AI Engineer
Kainovation Technologies
- End-to-End ML Pipelines: Architected and deployed a privacy-preserving AI platform, managing the full lifecycle from data ingestion to multi-LLM inference and secure CI/CD pipelines.
- MLOps & Automation: Built an enterprise-grade benchmarking framework to automate model testing and validation, significantly improving deployment readiness and evaluation reliability.
- Computer Vision & OCR: Developed high-throughput OCR and document intelligence pipelines, utilizing deep learning models to clean and extract data from complex, low-quality datasets (invoices, forms).
- Generative AI & NLP: Designed a Text-to-SQL system for ERP platforms, enabling non-technical users to query complex databases using natural language, bridging the gap between data and business teams.
- Model Deployment: Implemented a real-time PPE detection system using custom YOLO models, optimizing inference performance for edge deployment and automated safety monitoring.
- Multi-Agent Systems: Deployed a multi-agent backend for AR assistants, orchestrating tools for real-time interaction and voice processing.

Data Scientist
Kainovation Technologies
- Architected a production-ready ML pipeline for Customer Churn Prediction, integrating Kafka for real-time streaming, Airflow for orchestration, and MLflow for tracking.
- Built time-series forecasting models for agricultural commodity price prediction, supporting data-driven supply chain optimization.
- Implemented an AI-powered logistics optimization system using Google OR-Tools to solve complex VRP problems, improving route efficiency and fuel utilization.
- Engineered an NLP-driven sentiment analysis engine to perform real-time opinion mining on social media streams, identifying emerging market trends.

Machine Learning Engineer Intern
Kainovation Technologies
- Developed an analytics dashboard for a power plant, enabling real-time water quality monitoring, chemical dosing optimization, and predictive insights.
- Developed an algorithmic trading system leveraging multi-timeframe analysis, RSI, moving averages, and volatility indicators.
- Designed and deployed automated data scraping pipelines to aggregate historical market data, laying the foundation for high-frequency algorithmic trading models.
- Assisted in the development of optimization algorithms using Python, cleaning and structuring geospatial data for complex logistics and routing simulations.
- Performed exploratory data analysis (EDA) and feature engineering on industrial sensor data to support predictive maintenance modeling for power plant systems.
- Containerized machine learning workflows using Docker to standardize development environments and accelerate model testing cycles.
- Built interactive data visualization dashboards using Streamlit to monitor real-time model performance and dataset health.

Member
Institute of Mathematical Statistics (imstat.org)
- Full-time · Ohio, United States · Remote
- Member Institute of Mathematical Statistics (IMS) 2024 – 2025 maintained membership in this leading international scholarly society to stay current with the latest developments, theory, and applications in statistics and probability.

AI Engineer | Data Scientist
Freelancer
- Custom LLM & Agentic Workflows: Engineered specialized AI chatbots and RAG pipelines using LangChain, LangGraph, n8n and AutoGen.
- Multi-Agent Marketing System: Developed an autonomous multi-agent system to orchestrate complex marketing campaigns.
- Computer Vision: Architected automated image classification models for e-commerce platforms.
- End-to-End Data Pipelines: Designed scalable ML and Data Engineering pipelines on Azure, AWS, and GCP, facilitating real-time financial analytics and data-driven decision-making.
- Intelligent Document Processing: Implemented high-accuracy OCR solutions to automate data extraction from structured and unstructured documents, saving clients significant manual processing hours.
- Academic Research Support: Technical and data-driven assistance for Final Year and MBA research projects, including statistical analysis, data visualization and statistical modeling on primary and secondary research data.
- Large-Scale Data Extraction (Web Scraping): Built scalable data pipelines to collect and process large volumes of unstructured web data for competitive analysis and market research.
Technical Stack
AI, Machine Learning & Computer Vision







LLM, Gen AI, Multi-Agent Systems & Agentic Workflows












Databases





Development & Web




Cloud & Infrastructure





Data Science & Analysis


Data Engineering & Orchestration





Business Intelligence & Visualization


Selected Work
Marketing Multi-AI-Agent System
A hierarchical multi-agent orchestration framework built using the Google Agent Development Kit (ADK). It orchestrates specialized agents to handle research, copywriting, and visual creation for brand-aligned campaigns.
Customer Churn Prediction Pipeline
End-to-end production ML pipeline to predict customer churn. Features real-time streaming with Kafka, airflow orchestration, and MLflow experiment tracking.
Construction Document Automation
Automating extraction and generation of construction documents (WIR, NCR, RFI) using OCR and LLMs like Gemini. Streamlines method statement creation and inspection requests.
PPE Detection System
Real-time detection of Personal Protective Equipment (PPE) using YOLOv8/YOLOE. Identifies vests, helmets, gloves, and ensures safety compliance via video feeds.
Multimodal Chatbot for Annual Reports
Multimodal chatbot analyzing financial reports (text, tables, charts). Uses RAG with vector DBs to provide insights from PDFs.
Accomplishments
Honors, awards, and certifications earned throughout my journey.


Most Popular Researcher
Recognized as the Most Popular Researcher at the 24th ICTer International Conference.
Event Details

Machine Learning Hackathon: 2nd Runners-Up
Secured 2nd Runners-Up position at the International Conference in Data Science (ICDS) 2023 Machine Learning Hackathon.
Event Details
Generative AI Hackathon 2024: Nationwide
Finalist (Top 5) in a nationwide open category competition featuring both undergraduates and industry professionals.
Event Details
Google Cloud Arcade Champion
Achieved the Champion Milestone in Google Cloud's Arcade program in 2024.
Event Details
Finalist - Aurora AI Ideathon 2024
Reached the final stage of the Generative AI Hackathon: Aurora AI Ideathon 2024.
Event Details
International Generative AI Hackathon
Featured in the Top 25 International Ranking among global participants.
Event Details
Predicta 1.0: Machine Learning Hackathon
Achieved 4th place out of 80 teams in both Round 01 and Round 02 of the competition.
Event Details
Champion - Cambio Code & Design Challenge
Won first place in both the code and design challenges organized by Cambio Software at ICTer International Research Conference 2023.
Event DetailsGet In Touch
Have a project, research opportunity, or just want to say hi?
fathir.majeed@gmail.com
Location
Dubai, United Arab Emirates


