About Mohamed Sheriff
I build AI systems for organizations, on twenty years of running the systems businesses depend on.
The AI work is current. What makes it practical is everything underneath it: enterprise data, Oracle E-Business Suite, national telecommunications infrastructure, and executive responsibility for services people could not do without.
Professional foundation
Most AI advice comes from one side of the problem. This comes from both.
Today I help organizations decide where AI is worth applying, automate the workflows that are costing them time, make their data trustworthy enough to act on, and turn product ideas into specifications a team can actually build.
That work is grounded in two decades of enterprise delivery — business intelligence, ETL, data integration, Oracle E-Business Suite, analytics and reporting — with Accenture, Amdocs, BMC Software, DST Innovis and Hanover Compressor.
It is also grounded in something less common among AI consultants: executive accountability for national infrastructure. Leading Sierra Leone’s cable landing organization, chairing the assignment routing and restoration subcommittee of an international submarine cable consortium, and deploying the country’s first terrestrial fiber network teaches a particular respect for systems that must not fail.
That combination is the reason the recommendations are conservative about what technology can do and specific about what it takes to run.
How the foundation was built
Five stages, each of which changed how I approach the current work.
None of this is presented as a former career. Each stage is the reason a particular part of the AI work is done the way it is.
Software and databases
Learning what production actually means
Early work in software, databases, reporting and business intelligence established how applications and organizational data hold up under daily operational load — and how they fail.
Enterprise consulting
Reading systems built by other people
Oracle E-Business Suite, ETL, analytics and cross-module reporting with Accenture, Amdocs, BMC Software, DST Innovis and Hanover Compressor. Tracing transaction flows through enterprise data models is the same skill an AI opportunity assessment needs.
Telecommunications infrastructure
Building things that cannot be allowed to stop
Submarine cable construction, assignment routing and restoration, technical procurement, terrestrial fiber deployment and DWDM educational connectivity. Infrastructure work is where the difference between a working demonstration and a running service becomes obvious.
Executive leadership
Owning the consequences
Chief Executive Officer and Managing Director of Sierra Leone Cable Limited, with responsibility for governance, stakeholders, procurement, operational risk and service continuity. Technology decisions stop being technical when the organization answers for them publicly.
Founder and AI practice
Applying all of it
Through Anunnaki Technology Solutions LLC: AI strategy, automation, SaaS product planning, data, media systems, education and community platforms — built with the same discipline the infrastructure work demanded.
Working principles
A disciplined approach to technology and AI.
- 01
Begin with the problem
Technology selection should follow a clear understanding of the current process, stakeholders, constraints, risks, and desired outcome.
- 02
Make decisions visible
Requirements, workflows, assumptions, priorities, acceptance criteria, and architectural choices should be documented clearly.
- 03
Treat data as infrastructure
Automation and AI depend on data quality, ownership, definitions, access, structure, and responsible use.
- 04
Design for adoption
A technically capable system still needs understandable workflows, responsible ownership, training, support, and stakeholder participation.
Leadership perspective
What running critical infrastructure teaches you about AI projects.
Infrastructure has no tolerance for a system that works in a demonstration. It has to survive a bad day, an absent expert, a vendor dispute and a budget cycle. Executive responsibility for that means budgets, policies, people, risk, institutional priorities and public expectations were never someone else’s problem.
Applied to AI work, it produces a specific bias: name who will own the thing once it exists, be explicit about what it should not be trusted to do, and prefer the smaller system that can actually be operated over the larger one that cannot.
It also makes the conversation work in both directions — with the engineers who will build it and with the board that has to approve it.
Anunnaki Technology Solutions LLC
A company created to explore and build practical solutions.
Founded in 2023, Anunnaki Technology Solutions LLC provides the organizational foundation for Mohamed Sheriff’s consulting, research, product-planning, AI, automation, media, education, and community-technology initiatives.
AI and workflow automation
Assessing opportunities and planning practical systems that reduce repetitive work or improve service delivery.
SaaS and product planning
Turning early ideas into product charters, MVP scopes, requirements, roadmaps, workflows, and acceptance criteria.
Data and decision support
Improving the organization, quality, reporting, and usefulness of operational information.
Education, media, and community platforms
Developing structured initiatives that support learning, communication, cultural connection, and digital participation.