ZEISMO
Research & lab
Research direction / ZEISMO AI Engine

Engineer
Mohanad Osama Zahran

Artificial Intelligence Expert

Engineer Mohanad Osama Zahran leads the concept and research direction of JARVIS and System Brain at ZEISMO Research & Lab. His published work covers AI agents, decision systems, memory and e-commerce resilience.

Official author profile at ZEISMO Research & Lab.

Explore JARVIS research ↗Research enquiriesMohanad Osama Zahran on LinkedIn
Author reference

Published work
and research subjects.

ZEISMO Research & Lab research lead for JARVIS and System Brain, and author of Beyond One Market.

Research subjects: AI agents, System Brain architecture, Agent memory, Decision evaluation, E-commerce resilience.

  1. JARVIS & System Brain: Towards accountable machine decision-making and progressively autonomous departments

    Technical preprint · 28 Sep 2026

  2. Inside System Brain

    Engineering article · 4 Oct 2026

  3. Beyond One Market: Political Disruption, E-Commerce Resilience and Overseas Growth

    Research report · 5 Oct 2026

  4. Two Countries, One Point of Failure: Testing E-Commerce Resilience

    Practical methods article · 5 Oct 2026

  5. When Should an AI Agent Forget? Expiring Memory Before It Becomes a Bad Decision

    Engineering methods article · 5 Oct 2026

  6. After the Last Human: Could AI Keep Working?

    Engineering perspective · 5 Oct 2026

  7. When Should an AI Employee Decide, Ask for Help, or Stop?

    Computational pilot · 7 Oct 2026

  8. After the Restart: Can an AI Employee Remember, Recover and Resume Work Safely?

    Computational pilot · 9 Oct 2026

Use the full name Mohanad Osama Zahran and the exact title, version and source link when attributing this work. Each publication states its methods and evidence scope.

From AI engineering
to decisions we can examine.

Mohanad Osama Zahran leads the concept and research direction of JARVIS and System Brain at ZEISMO Research & Lab. His work focuses on AI model training and adaptation for ZEISMO AI Engine, coordinated agents and the practical design of AI-operated workflows.

The programme investigates how an English-speaking interface, specialised software cells and useful memory can support a gradual transition of defined routine tasks to AI decision systems.

Professional title and ZEISMO AI Engine background supplied by Mohanad Osama Zahran. The research report records the evidence and scope of its own experiments separately.

01 / ENGINEERING

ZEISMO AI Engine

Model training and adaptation, role-specific instructions and practical AI system development.

Author-stated engineering focus
02 / RESEARCH

JARVIS & System Brain

English-speaking interaction, fifteen-cell prototype inspection, agent coordination and inspectable memory.

Read the technical report ↗
03 / EVALUATION

Evidence before autonomy

Reproducible experiments, explicit decision boundaries and evaluation of routine departmental workflows.

Explore the questionnaire ↗
Featured technical report

JARVIS &
System Brain.

Towards accountable machine decision-making and progressively autonomous departments.

72 pages connecting architecture, prototype inspection, controlled synthetic experiments and a proposed evaluation programme. The report carries Mohanad Osama Zahran’s authorised typed attribution and describes the contribution of AI-assisted preparation.

Open report PDF ↓Methods & supporting files
A conversation about useful intelligence

Bring a question.
Challenge the method.

Researchers, engineers and institutions can engage around agent reliability, meaningful evaluation and the evidence needed to delegate a task.

Connect with the lab ↗Research forum ↗
Engineering article / 4 October 2026

Inside System Brain

By Mohanad Osama Zahran. Explore how role-specific agents, tools and memory can coordinate accountable decisions.

Read the illustrated article ↗
Research profile

Questions, work
and evidence.

Who is Mohanad Osama Zahran?

Engineer Mohanad Osama Zahran leads the concept and research direction of JARVIS and System Brain at ZEISMO Research & Lab. His stated engineering focus is ZEISMO AI Engine and role-specific AI workflows.

Where can readers examine the published work?

The JARVIS & System Brain technical preprint contains methods, prototype tests, synthetic experiments and supporting files. Inside System Brain explains the architecture and distinguishes the inspected prototype from proposed capabilities.

Which work is still being evaluated?

The follow-on study asks when an AI employee should decide, ask for help or stop. The computational pilot is published, comparing fixed rules, one model call, a System Brain pattern and self-review. Its original blinded human explanation-review endpoint remains unperformed.

How should this research be attributed?

Use the full author name, Mohanad Osama Zahran, and cite the exact work, version and source link. Check the evidence scope and the applicable reuse terms before republishing.

AI employees and decision systemsFollow published work through RSS
Explore in depth

Research, engineering and published work.

01 / AUTHOR GUIDE

Research direction

The questions behind JARVIS, System Brain and the transition from AI recommendations to accountable decisions.

Explore research direction ↗
02 / AUTHOR GUIDE

Engineering approach

Role-specific software cells, scoped tools and inspectable memory in the ZEISMO AI Engine research programme.

Explore engineering approach ↗
03 / AUTHOR GUIDE

Publications guide

A guide to published work, supporting artifacts and the next study from Engineer Mohanad Osama Zahran.

Explore publications guide ↗
Research report / 5 October 2026

Beyond One Market

Political Disruption, E-Commerce Resilience and Overseas Growth. By Mohanad Osama Zahran. A 34-page evidence review and planning framework with reproducible synthetic scenarios.

Read the report ↗
New methods articles / 5 October 2026

Dependencies, memory and better decisions.

Two Countries, One Point of Failure

Mohanad Osama Zahran explains how shared warehouses, payments and operating access can undermine overseas diversification, with a practical dependency test.

Read the article ↗

When Should an AI Agent Forget?

Mohanad Osama Zahran proposes a System Brain memory lifecycle for stale evidence, source revalidation and accountable act, ask or stop decisions.

Read the article ↗
Computational pilot / 7 October 2026

Decide, ask for help, or stop?

Mohanad Osama Zahran presents a controlled comparison on 36 synthetic evaluation cases, with recorded model outputs, reproduced calculations and a ZEISMO Lab verification record.

Read the pilot and evidence ↗
Computational pilot / 9 October 2026

After the restart.

Mohanad Osama Zahran examines how an AI employee handles saved memory, changed permissions and completed work. Compare four memory conditions, inspect response-format failures and download the reproducible record.

Read the study and evidence ↗
ZEISMO research practices

Evidence you can inspect.

Follow each marker to the method, record or policy behind it.

Versioned recordsIdentifiers and integrity checks Documented methodsProtocols and reproducibility tools Inspectable evidenceMethods, aggregates and fingerprints Accountable publishingAuthorship and correction policies

ZEISMO practice markers. Verification scope is recorded with each publication.