About
Senior GenAI/NLP Engineer and Tech Lead with 8+ years of experience building LLM-powered assistants, multi-agent systems, and knowledge-graph products. Currently leading an AI team delivering enterprise workflows using multi-agent and NLQ-to-SQL solutions. Strong background in RAG, GraphRAG, and knowledge alignment, with hands-on delivery across Python/Java stacks, Neo4j, and production-grade LLM systems.
Experience
Dec 2025 – Present
Gen AI Team Lead Stealth
- Building an enterprise multi-agent AI system for business process intelligence using Microsoft AutoGen
- Developing a NLQ-to-SQL pipeline using DSPy with SQL validation to safely execute AI-generated queries
- Architected real-time agent communication via gRPC bidirectional streaming
Dec 2022 – Dec 2025
Lead NLP Engineer Implicit (formerly Agolo)
- Leading the Answers Team to develop an advanced customer support chatbot using GraphRAG, Neo4j knowledge graphs, and a self-hosted LLaMA-based LLM, enabling automated responses to product support queries
- Designed and implemented a German-language Q&A system powered by Neo4j + GPT-4, sourcing answers from complex product manuals
- Led R&D efforts in hallucination detection via FactAlign, a knowledge graph-based alignment technique published at TrustNLP 2024
- Architected and fine-tuned domain-specific LLaMA models for summarization and generation tasks in enterprise use cases
- Integrated PDF parsing using YOLOv7 and rule-based extraction for structured information retrieval
Dec 2021 – Dec 2022
Senior NLP Research Engineer Implicit (formerly Agolo)
- Spearheaded the Summarization Team, focusing on PDFs, ESG reports, and news summarization through RAG-based and entity-centric KG summarizers
- Directed efforts in fine-tuning LLMs (LLaMA family) for custom summarization and text generation pipelines
- Collaborated cross-functionally with product teams to embed summarization capabilities into customer-facing applications
Dec 2020 – Dec 2021
NLP Research Engineer Implicit (formerly Agolo)
- Developed a PDF parsing system combining rule-based extraction and YOLOv7, enabling accurate retrieval of structured text from complex document layouts
- Built and evaluated summarization pipelines for PDFs, ESG reports, and news articles, contributing to the foundation of Agolo’s entity-centric and RAG-based summarizers
- Contributed to early LLM experimentation and supported fine-tuning workflows for domain-specific summarization and QA systems
Dec 2018 – Dec 2020
Research Assistant Biomedical & Neuro Engineering Lab
- Built neural spiking feature extraction and classification pipelines using SVM and PCA for ALS biomarkers
- Achieved ~81% accuracy on ALS-model mice using SVM and ensemble methods
Sep 2016 – Mar 2020
Teaching Assistant Ain Shams University
- TA and grader for Neural Networks, Data Structures & Algorithms, Computer Vision, and Digital Image Processing
Publications
FactAlign: Fact-level hallucination detection and classification through knowledge graph alignment
TrustNLP Workshop @ NAACL 2024 Spiking motoneurons effective connectivity as an early electrophysiological biomarker of ALS
Neuroscience Meeting Planner, SfN 2021 A classification approach to recognize the firing of spinal motoneurons in ALS
IEEE EMBC 2020
TrustNLP Workshop @ NAACL 2024 Spiking motoneurons effective connectivity as an early electrophysiological biomarker of ALS
Neuroscience Meeting Planner, SfN 2021 A classification approach to recognize the firing of spinal motoneurons in ALS
IEEE EMBC 2020
Education
M.Sc. in Computer Engineering
Ain Shams University · 2016–2021 · GPA 3.5
Thesis: "Extraction of Electrical Markers for Motor Neuron Disease Using Machine Learning Methods"
B.Sc. in Computer Engineering
Ain Shams University · 2011–2016 · GPA 3.5
Graduation project: AUTOSAR Ethernet Opponent
Skills
Languages
AI / ML
Infrastructure
Spoken