GENCITY® Institutional programmes
Proposal prepared for discussion with IT Park Uzbekistan
Institutional partnership proposal · Uzbekistan

Applied AI Factory & Engineering Academy for IT Park Uzbekistan

A practical Phase 0 execution node designed to turn national AI ambition into trained engineers, locally deployable AI systems and commercial startup prototypes.

The proposal is simple: establish a small, operational AI factory inside the IT Park ecosystem before larger national-scale compute comes fully online. The node would combine local NVIDIA-class compute, hands-on engineering training, startup production and controlled sovereign AI workloads.

Initial scope: 1 × NVIDIA DGX Station + 1 × NVIDIA DGX Spark Initial cohort: 20–50 engineers / startup teams Programme horizon: 6–12 months

Why this fits Uzbekistan now

  1. Uzbekistan is already evaluating an AI Center of Excellence based on modern NVIDIA technologies.
    Public plans also reference faculty training, NVIDIA DLI, university AI infrastructure and approximately 1 MW of compute capacity across education/research and government/industry use cases.
  2. IT Park is already developing the human-capital layer.
    AI Leaders 2026 combined strategic AI education, applied workshops and agent-development content for executives and institutions.
  3. GPU access for startups is already an active policy discussion.
    IT Park and resident companies have publicly discussed a GPU Cluster project and access to compute resources for domestic startups.
This project is not a competing national cluster. It is a small execution layer that can produce operators, prototypes, operating procedures and measurable evidence before larger infrastructure is scaled. In institutional terms: build the human and application pipeline alongside the compute pipeline.

Phase 0: a working node, not a hardware showcase

The objective is to create a controlled environment where engineers and startups learn by building real systems on the same class of local AI infrastructure they may later operate at larger scale.

Primary engineering node NVIDIA DGX Station, powered by the GB300 Grace Blackwell Ultra Desktop Superchip. Intended for advanced local inference, model development, fine-tuning, agents and data-science workloads.
Developer / startup node NVIDIA DGX Spark. Intended for prototyping, local inference and practical fine-tuning workflows, including parameter-efficient methods.
Operating principle Local execution by default for sensitive or policy-restricted data; cloud and external APIs used only where permitted and useful.
Procurement approach Jointly explore strategic hardware contribution, evaluation allocation, preferential procurement, co-funded deployment or another mechanism acceptable to NVIDIA and IT Park. No assumption of free hardware.

Final hardware, software, licensing and delivery are subject to vendor availability, export controls, local procurement rules and separate agreements.

What participants would build

Applied Agent Engineering

Python, approved LLM APIs, RAG, document ingestion, embeddings, vector search, enterprise knowledge systems, agent workflows and deployment.

Local AI Engineering

Local inference, 3B–9B model workflows, LoRA / QLoRA / SFT, synthetic datasets, evaluation, optimization, Jupyter and the NVIDIA software stack.

Commercialization

Every team exits with an auditable artifact: a working agent, RAG application, fine-tuned model, enterprise proof of concept or commercial prototype.

Success is measured by production evidence, not attendance. The programme is designed to create a pipeline of people who can operate AI infrastructure and a pipeline of applications worth running on it.

Measurable 6–12 month outcomes

Human capability

Initial cohort of 20–50 engineers, founders and startup teams trained through practical engineering tasks and deployment work.

Applied portfolio

A portfolio of working AI agents, RAG systems, local model workflows and sector-specific prototypes suitable for technical review.

Institutional use cases

Selection of controlled government, enterprise, education or industry pilots where local AI execution provides a clear data-governance or operational benefit.

Scale readiness

An operating playbook, training baseline and application pipeline that can inform future expansion into larger national AI infrastructure.

Proposed institutional roles

IT Park Uzbekistan

Institutional host; nominate a project owner; provide the local programme environment, participant access and coordination; support alignment with relevant Ministry and national AI initiatives.

GENCITY

Provide programme architecture, engineering curriculum, mentor enablement, deployment methodology, local-AI/RAG/agent workflows, evaluation standards and commercialization structure.

NVIDIA

Subject to separate NVIDIA approval: explore technology infrastructure participation, technical enablement, appropriate software access and DLI alignment. NVIDIA participation is not represented as agreed or guaranteed by this proposal.

International delivery ecosystem

GENCITY is supported by affiliated companies under common management across the United States, United Kingdom and United Arab Emirates.

HYBT Technologies, Inc.

Delaware, United States. Operator of the GENCITY platform and AI infrastructure / applied engineering arm. HYBT Technologies, Inc. is an NVIDIA Inception member and an AI for Good initiative participant.

Programme participation is presented as an organisational credential and does not imply endorsement, sponsorship or approval of this proposal.

AISC TECHNOLOGIES LTD

England & Wales · Company No. 16665302. UK software, web-portal and education company; owner of the registered GENCITY® trade mark and the group’s UK intellectual-property base.

AL SAFI FZE

Umm Al Quwain, United Arab Emirates. UAE education, EdTech and innovation activity, including participation in the Ras Al Khaimah Teachers Network (RAKTN), which is managed by the Sheikh Saud bin Saqr Al Qasimi Foundation for Policy Research.

Decision requested from IT Park

  • Confirm interest in developing a joint Phase 0 concept note.
  • Nominate an IT Park project owner for technical and institutional scoping.
  • Agree the first cohort profile, host environment and target use cases.
  • Subject to internal approval, prepare a joint technology-engagement route with NVIDIA and relevant government stakeholders.

Public context

The proposal is intentionally aligned with publicly stated initiatives rather than presented as a new national policy.

  1. IT Park Uzbekistan — The Republic of Uzbekistan Launches Major AI Initiative with NVIDIA
  2. IT Park Uzbekistan — AI Leaders 2026 concludes in Tashkent
  3. IT Park Uzbekistan — GPU Cluster and industry growth initiatives
  4. NVIDIA — DGX Station and NVIDIA — DGX Spark
  5. Companies House — AISC TECHNOLOGIES LTD
  6. UKIPO — GENCITY trade mark UK00004280109
  7. Ras Al Khaimah Teachers Network (RAKTN)
  8. AI for Good — global initiative
  9. HYBT Technologies, Inc. — SYNQAGI StateDelta Edge public repository
  10. HYBT Technologies, Inc. — Proportional Cost and Energy Allocation for Local Sovereign AI Nodes (DOI 10.5281/zenodo.20569427)

Open a working discussion

For IT Park Uzbekistan, Ministry, university or institutional stakeholders who want to evaluate the Phase 0 model, the next step is a focused scoping discussion covering host environment, cohort, use cases, governance and technology engagement.

Open institutional enquiry