About
I'm a First Class Computer Science graduate from Brunel University London, currently working as a Junior Data Scientist at EmergeIQ Ltd. My passion lies in building AI systems that solve real-world problems—from document intelligence and compliance verification to automated video analysis.
What drives me is the challenge of transforming messy, unstructured data into functional, reliable AI systems. I don't just build models; I engineer complete solutions with proper validation, security, and production-grade architecture. Every system I develop emphasizes thorough testing, OWASP LLM Top-10 compliance, and human-in-the-loop validation to ensure alignment with business requirements.
Beyond the technical work, I'm fascinated by the intersection of AI and practical problem-solving. Whether it's designing RAG pipelines for semantic search, implementing computer vision for video annotation, or developing agent-based workflow automation, I approach each project with a focus on reliability, security, and measurable impact.
What I Do
- IOffering
LLM Systems Engineering
Design and build production-grade LLM applications with advanced prompt engineering, RAG architectures, and agent-based systems.
- IIOffering
AI Security & Compliance
Implement comprehensive security measures to protect AI systems from vulnerabilities and ensure regulatory compliance.
- IIIOffering
Computer Vision Solutions
Develop CV pipelines for video analysis, object detection, and automated annotation using modern frameworks.
- IVOffering
Backend Development
Build robust, scalable API backends with comprehensive authentication, validation, and monitoring.
- VOffering
Document Intelligence
Extract, analyze, and structure information from unstructured documents using AI-powered pipelines.
- VIOffering
System Integration
Connect AI capabilities with existing systems through well-designed APIs and integration patterns.
Approach
My approach to AI engineering is grounded in three core principles:
**1. Security First**: Every AI system I build incorporates security from the ground up—not as an afterthought. This means implementing prompt injection mitigation, input validation, output filtering, and achieving full OWASP LLM Top-10 compliance.
**2. Validation & Reliability**: AI outputs need systematic validation. I implement comprehensive testing frameworks to identify data gaps, contradictions, and logical inconsistencies before they reach production.
**3. Human-Centered Design**: AI should augment human capabilities, not replace human judgment. I design systems with controllability and human-in-the-loop validation, ensuring alignment with business requirements and user needs.
Interests
Outside the terminal: ai research & innovation, open source contribution, problem solving, continuous learning.
Staying current with the latest developments in LLMs, RAG architectures, and AI security. I regularly explore new frameworks and methodologies to improve my craft.