About Nick Tabbal, founder of Agentic Consulting.

I have been building machine intelligence into production systems for over thirty years. The technology changed every few years. The discipline did not.

I started with statistical models. I moved into machine-learning-based predictive models. I ran natural language processing in production in its early days and fine-tuned it with human analysts before fine-tuning was a feature you could buy. At BlackSky I worked with computer vision at scale on satellite imagery. At Agentic Consulting the work is LLM-based and agentic. Every wave looked new. The question underneath it never changed. Will this survive contact with real data and the people who have to use it.

That question is why I founded the firm. The pattern I kept seeing inside operating companies was the same. Leadership made the right strategic call. The technology investment was sound. The daily work did not change. The gap between intent and workflow was where the value disappeared, and the people responsible for closing it were already running the company. I built Agentic Consulting to install the operating layer that closes that gap. The work is operator-level. It does not produce strategy decks. It produces systems that run after I leave.

I scaled Resonate as its second employee, applying psychographic-targeting algorithms across 230 million users. I co-founded the AdAdvisor division at TARGUSinfo and built privacy-friendly targeting before privacy was a regulatory mandate. I pioneered machine-learning analytics at Bulletin Intelligence that enabled the exit to Cision and delivered daily intelligence briefings to the White House and Cabinet agencies. I ran a global client success organization of more than 100 people at Comscore managing 200 to 240 million dollars in ARR. I built the customer-facing organization at BlackSky from nothing and scaled it from zero to 50 million dollars in ARR in under four years.

The lesson across all of it is that the model is rarely the constraint. The constraint is whether the organization can operate the result. That is the work I do now, and it is also the foundation for the second kind of work we do, building AI and data products for companies that own the raw material.

I do not sell training, implementation, or strategy to a company that has no operating layer ownership. If the work will not produce the return the client needs, I disengage rather than book the revenue. This is the boundary that protects both the client and the practice.

Every engagement starts with a diagnosis. The 30-minute call is the first step, and the substance of the call is identifying where the operating layer gap actually is. If a fractional engagement fits, we discuss that. If workforce enablement or the build phase fits, we discuss those. If no engagement I offer will produce the return needed, I say so directly.

I work as a solo practitioner with a long-standing implementation partnership with IgniteIQ. The relationship spans 30 years and operates on full joint delivery accountability. When an engagement requires platform implementation, IgniteIQ delivers the build alongside me. The client gets a single point of accountability and a delivery partnership that has been operating together for three decades.

I hold the MIT Professional Education certificate in Applied Agentic AI for Organizational Transformation. I hold a Bachelor of Engineering with Honors from Imperial College, University of London. I formerly held a Top Secret clearance with SCI access and a CI polygraph. I am a named inventor on US Patent 10,475,047 for ad targeting, with three additional applications.

The method I use draws on the AI Maturity Cycle developed by Dr. Abel Sanchez, the Crawl Walk Run scaling model, the NIST cybersecurity framework, the ADKAR and Kotter change management approaches, and the OECD AI principles. Every deployment is staged. Every stage has reliability thresholds. Security and governance are foundations, not phases. The frameworks are safety rails. The outcome is measurable productivity within 90 days.

The methodology page describes the full framework.