CAIRO, EGYPT · REMOTE WORLDWIDE

Mohammed Ehab ElNomany

Data EngineerAI EngineerData ScientistBusiness AnalystData Analyst

I turn complex data and business problems into reliable systems, models, analytics and products.

Explore selected work Discuss an opportunity
Data EngineerBusiness Analyst Team LeadPrivate Tutor · Computer Science & DataData / ML InternML InternCertified Data Engineer AssociateFoundation & AdvancedBSc Computer Science · Data Science MajorData Science & AI Scholarship

Six projects. One standard.

Each project opens to a full case study with inspectable evidence.

Each project is backed by a full case study and inspectable evidence.

PRESAIRA

Sports forecasting, on the record.

CHAMPIONPREDICTED
FINALEXACT MATCHUP
SEMIFINALISTS3 OF 4
WORLD CUP 2026 logo
WORLD CUP 2026PROOF RECORD
UCL 2026/27 logo
UCL 2026/27ACTIVE
FORMULA 1 logo
FORMULA 1NEXT
NBA logo
NBANEXT
Forecast calibration after the 2026 World CupFive observed win-rate points plotted against predicted probability from committed public Presaira calibration evidence. The dashed diagonal indicates perfect calibration.0.00.00.20.20.40.40.60.60.80.81.01.0Predicted probabilityObserved win rate
104/ 104World Cup matches scored
UCL 2026/27 ACTIVEFORMULA 1 + NBA COMING SOON
View case study

APPROACH

I turn messy reality into reliable systems.

I make unclear problems legible, reduce ambiguity, then build the smallest system that can carry the job.

  1. 01

    Messy reality

    Fragmented, inconsistent and unclear.

    Fragmented data
    Conflicting definitions
    Missing context
    Incomplete evidence
    1. 02

      Expose the truth

      Surface evidence, constraints and unknowns.

      EvidenceConstraintsUnknowns
    2. 03

      Reduce ambiguity

      Turn fuzzy questions into explicit mappings and decisions.

    3. 04

      Build the system

      Create the smallest reliable system that works in practice.

  3. 05

    Reliable outcomes

    Validated, usable and ready for real use.

    • Validated migration
    • Decision-ready analytics
    • Evaluated model
    • Governed AI workflow
    • Usable product

What I’m thinking through.

Ideas, experiments, and everything that piques my curiosity as I navigate my career.

Data · Forecasting9 min read~8 min listen

When should you trust a probabilistic forecast?

A practical test for knowing when a probabilistic forecast deserves trust.

Data · Computer vision8 min read~8 min listen

Why accuracy alone is not enough for oil-spill detection

Why rare oil pixels make accuracy a weak headline metric.

AI · AI agents9 min read~8 min listen

What should an AI agent do when the evidence is missing?

How agents should handle missing evidence without inventing certainty.

Choose the right conversation.

Hiring for data, AI or product?

Start with the work, then reach me with the role and problem space.

EmailLinkedInWork

Have a system or product problem?

Start with the service context, then reach me with the system and outcome.

EmailLinkedInServices

Have a question or idea?

If it does not fit a role or project brief, email me or start a conversation on LinkedIn.

EmailLinkedIn