AK

Building agent memory & infrastructure — one system at a time.

About

I'm an AI infrastructure engineer focused on the problem of agent memory — how AI agents remember, share, and curate knowledge across sessions and across teams of agents.

I'm the founder of Quantify Labs Ltd. (UK) and the creator of Aegis Memory, an open-source persistent memory system for multi-agent AI. I also built mcp-parapet, security middleware for the Model Context Protocol targeting EU AI Act compliance.

Before AI infrastructure, I spent 10+ years in data engineering and business intelligence, working with SQL, Python, Azure, Databricks, and enterprise data pipelines. That background shaped how I think about agent systems — reliability, observability, and failing gracefully matter more than demos that work once.

I hold an MSc in Big Data Analytics from Sheffield Hallam University and I'm building toward a funded PhD (2027–28 target) researching multi-agent memory systems and collective knowledge discovery — specifically, whether quality-curated memory can make multi-agent simulation reliable enough to trust.

What I'm working on now

Learning AI agent infrastructure systematically through an 80-lesson curriculum covering LangChain, LangGraph, CrewAI, production deployment, and observability. Each lesson becomes a blog post here — not a learning diary, but a reference that explains what I built, what I learned, and what the documentation doesn't tell you.

In parallel: running a Polymarket trading bot (paper trading phase), preparing mcp-parapet for open-source launch, and writing a three-paper research arc for PhD applications.

Certifications

Timeline

2026
mcp-parapet v0.1.0 code-complete
2,778 lines of Python, 63 tests. Security middleware for MCP with content scanning, trust boundaries, HMAC integrity, and audit trails.
2026
Polymarket trading bot live (paper trading)
Hybrid strategy bot running on Oracle Cloud AMD VM. Weather and arbitrage strategies with positive early equity.
2025–26
Aegis Memory v1.3.0
Persistent memory for multi-agent systems. Scope-aware access, semantic search with pgvector, ACE patterns, CrewAI and LangGraph integrations.
2024
Founded Quantify Labs Ltd.
UK-registered company building AI agent infrastructure and developer tools.
2023
MSc Big Data Analytics
Sheffield Hallam University.

Get in touch

Best way to reach me is LinkedIn. I also post on DEV.to.