Skip to content
View AviElbaz7's full-sized avatar

Highlights

  • Pro

Block or report AviElbaz7

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AviElbaz7/README.md

Hi, I'm Avi Elbaz 👋

GenAI Engineer | Agentic Infrastructure & Multi-Agent (A2A) Orchestration

🌍 Based in Israel  •  🤖 Building the infrastructure that powers autonomous AI systems


About Me

  • 🧠 GenAI Engineer with 2+ years of experience specializing in core agentic infrastructure and multi-agent (A2A) orchestration.
  • 🏗️ I build secure agent environments, scalable data pipelines, and the orchestration layers that power autonomous systems.
  • 💼 R&D Software Engineer (GenAI & Algorithms) @ Amdocs.
  • 🎓 B.Sc. Data Science & Engineering, Ben-Gurion University (GPA 88).
  • 🤝 Open to collaborating on agentic platforms, LLM orchestration, and data-intensive systems.

🚀 What I Work On

  • 🔌 MCP Gateway & Registry - deployed an open-source MCP Gateway (React UI, Nginx + FastAPI, OIDC/OAuth2, RBAC) as the central agent registry, onboarding 30+ MCP servers with FAISS semantic discovery and YARA/LLM security gating.
  • Reactive A2A Orchestration - built an asyncio reactive DAG coordinator that fires each node the moment its parents complete, bypassing superstep barriers for a 20–30% reduction in configuration time.
  • 🔗 LLM I/O Aggregation - engineered a layer that reconciles mismatched agent outputs into schema-conformant inputs, resolving ~90% of input-mismatch failures in multi-agent convergence flows.
  • 📊 Agent Observability - end-to-end pipeline with RabbitMQ, Logstash, Elasticsearch, and Grafana for per-step visibility into LLM calls, MCP tools, A2A interactions, and token costs - cutting MTTR by ~60%.
  • 🌊 Real-Time ETL at Scale - architected pipelines using AWS Kinesis, Lambda, Glue, and PySpark, processing millions of events daily for live analytics and ML inference.

💻 Tech Stack

Languages

Python Bash Java C++ C

AI & GenAI

LangChain LangGraph MCP A2A RAG Amazon SageMaker

Data & Cloud

AWS Apache Spark Snowflake pgvector Pandas Apache NiFi Microsoft Fabric

Backend & Infra

FastAPI GraphQL REST Docker PostgreSQL MongoDB Elasticsearch Grafana


📈 GitHub Stats


🌐 Connect With Me

LinkedIn Email GitHub

Pinned Loading

  1. Solar-Panels-Analysis-Project Solar-Panels-Analysis-Project Public

    This repository contains the analysis and findings for our team project in SISE2601 - Data Analysis with R

    1

  2. Figure-vs-Monster-Cpp Figure-vs-Monster-Cpp Public

    Dynamic card-based game in C++, employing polymorphism for strategic battles between monsters and heroes. Implemented unique abilities and damage calculations, special powers for each card, the her…

    C++ 1

  3. Pokemons-C Pokemons-C Public

    Developed a versatile Pokemon terminal in C leveraging object-oriented programming principles. The program reads data from a file and creates based on the file Pokemon Structures the do several com…

    C 1

  4. Track-Files-Bash Track-Files-Bash Public

    Developed a folder tracking script using Bash under a VM to monitor changes within specified directories. The script, designed to enhance productivity and organization, efficiently identifies updat…

    Shell 1

  5. Crawler-For-X Crawler-For-X Public

    Developed an X (Twitter) Selenium scraper to extract timelines, date-filtered hashtags, and advanced follower analytics. The repository was adopted by SCE and utilized by all students in the domain.

    Python 1

  6. k-means-pyspark k-means-pyspark Public

    This repository contains a Jupyter Notebook that demonstrates the implementation of the K-Means clustering algorithm using PySpark.

    Jupyter Notebook 1