AI & Data · RP-AI-0018

Manar Ayman

MidAI Agent Engineer

AI agent engineer building multi-agent LLM platforms on Claude and Gemini, grounded in data engineering

  • Identity verified
  • Remote Pro Ai team member · under contract
Add to squadAll team profilesAI & Data · Game Development

Availability on request

  • 2yearsprofessional experience
  • MidSeniority level
  • Agentic AI systemsPrimary stack
  • CairoBased in · Egypt · GMT+2

Manar Ayman

Profile summary

Agentic AI engineer who single-handedly designed and built an 8-agent orchestration platform on Claude and Gemini, covering routing, RAG, media approval and a dedicated cross-agent handoff layer, and shipped it to production beta for a US marketing agency. Pairs LLM orchestration with hands-on AWS and GCP data engineering, including a Redshift Serverless reconfiguration that cut costs ~90%. Former ITI data-management teaching assistant with 400+ teaching hours in SQL, Power BI and Python; a chemistry and QC background drives a strict habit of tracing every number to its source.

  • Languages

    • ArabicNative
    • EnglishProfessional working proficiency

  • Sectors

    • Media
    • Education

Manar Ayman

3 roles

Experience

Most recent role first, with outcomes as stated in the CV.

  1. AI Data Engineer

    Creativo AI

    – Remote (USA)

    • Designed and built, as sole engineer, an 8-agent orchestration platform (routing, media approval, media search, email, docs/RAG, onboarding, caption generation and cross-agent handoff) on Claude and Gemini, shipped to production beta.
    • Replaced an early LangChain implementation with a custom orchestration framework once routing and handoff requirements outgrew the library.
    • Built an event-driven onboarding agent that assembles client brand books through conversation and syncs the extracted data via Firestore, Amazon S3 and Redshift to the client dashboard.
    • Cut Redshift Serverless costs ~90% (from ~$4,000 to under $300 per month) with a teammate by diagnosing a billing spike and reconfiguring workload capacity.
    • Eliminated mid-pipeline failures by co-migrating a Lambda ingestion pipeline to AWS Step Functions after it hit Lambda’s 15-minute execution limit.

  2. Data Management Teaching Assistant & Supervisor

    Information Technology Institute (ITI)

    – Cairo, Egypt

    • Delivered 400+ teaching hours across SQL optimisation, Power BI, Tableau, Python and database systems, including courses designed from scratch to match industry expectations.
    • Rebuilt the core Database course with a rewritten curriculum, engine-level query-tuning content and hands-on labs built around real-world, messy data.

  3. Quality Engineer

    My Way

    – Obour, Egypt

    • Ran structured quality-control analysis and maintained operational data records.

Manar Ayman

Selected projects

Shipped titles and products, with the role played on each.

  • Multi-Agent Customer Orchestration System

    Architect and engineer

    Composable, tool-callable agents instead of one large chatbot: a router handles first-pass classification and a dedicated handoff agent re-routes unclear messages, so users move between onboarding and media search without repeating themselves. Session state persists across logins, and brand-book fields are written only once actually extracted.

    • Claude
    • Gemini
    • LangChain
    • Firestore
    • Amazon S3
    • Amazon Redshift

  • Redshift Cost Reduction & Pipeline Reliability

    Engineer (with a teammate)

    Root-caused a billing spike to oversized workload capacity and reconfigured it to 2–⁠4 RPU, cutting monthly cost from roughly $4,000 to under $300 with no drop in query performance; moved ingestion to Step Functions so every step completes or fails visibly.

    • Amazon Redshift Serverless
    • AWS Lambda
    • AWS Step Functions

Manar Ayman

8 core skills

Skills

Core skills first, then the full toolkit by theme.

Core skills

  • Agentic AI systems
  • Multi-agent LLM orchestration (Claude, Gemini)
  • RAG & retrieval design
  • Tool/function calling
  • Python
  • Advanced SQL
  • AWS (Redshift Serverless, Step Functions, Lambda, S3)
  • GCP (Vertex AI, Firestore)

  • AI & generative AI

    • Agentic AI systems
    • Multi-step LLM orchestration
    • Claude
    • Gemini
    • RAG
    • Vector databases & embeddings
    • Function/tool calling
    • LangChain

  • Cloud – AWS

    • Amazon Redshift Serverless
    • AWS Lambda
    • Amazon S3
    • AWS Step Functions
    • Amazon QuickSight
    • IAM fundamentals

  • Cloud – GCP

    • Vertex AI
    • Firestore
    • Cloud Scheduling
    • ML model deployment

  • Data engineering

    • Data modelling (star/snowflake/galaxy)
    • ETL/ELT design & optimisation
    • Event-driven processing
    • Query performance tuning
    • Stored procedures
    • Apache Spark
    • Hadoop & Hive

  • Databases

    • Amazon Redshift
    • PostgreSQL
    • MySQL
    • Oracle (PL/SQL)
    • DynamoDB
    • MongoDB
    • Firestore

  • Languages & tools

    • Python
    • Advanced SQL (window functions)
    • PL/SQL
    • Java
    • Docker
    • Git

  • BI & analytics

    • Power BI
    • Tableau
    • QuickSight
    • Funnel analysis
    • KPI dashboards

Manar Ayman

Education and certifications

Degrees, certifications and awards on record.

Education

  • Bachelor of Science (Very Good with honours)Faculty of Science, Al-Azhar University · 2019
  • Intensive Training Program – Data Integration & Visualisation (Grade A)Information Technology Institute (ITI) · 2023

Certifications

  • Microsoft Certified: Power BI Data Analyst AssociateMicrosoft
  • Python 3 Programming SpecializationUniversity of Michigan
  • Lean Six Sigma Green Belt
  • IELTSBritish Council
  • Introduction to Git and GitHubGoogle

Manar Ayman

Usually works with

Roles this profile is usually paired with in a squad.

Build this squad

AI & Data

Bring this profile into your squad

Add the role and level to the Squad Builder; availability is confirmed with your quote.

Add to squadAll team profiles

Availability on request