AERésumé PDF ↓

Cairo, Egypt · Open to opportunities

AhmedEladl

AI engineer building voice AI and agent systems that scale in production.

See the workSend a message
  • 3+years exp
  • 1k+downloads
  • 8+papers
  • 2pypi packages

About

Hi there!

Ahmed Eladl
Ahmed Eladl, off duty

research mode

AI Engineer with 3+ years of experience building and deploying conversational and voice AI systems, from real-time voice agents to enterprise bots across SIP and WhatsApp Business channels. Skilled in LLMs, RAG pipelines, and MCP and tool-use integrations, with hands-on experience fine-tuning Arabic dialects TTS and ASR models and deploying low-latency inference on cloud-native infrastructure.

3+ years shipping production AI: Architected and deployed end-to-end conversational voice systems across SIP and WhatsApp Business, enterprise multimodal RAG, and MCP tool-use agents. Hands-on ownership taking complex neural models from research into scalable, high-reliability production on GCP with FastAPI, Kubernetes, KServe, and vLLM.

High-impact open source: Creator of ASRX (universal forced alignment & diarization engine on PyPI) and open-source Arabic speech models powering 1,000+ downloads across the global Hugging Face community.

Research rigor & academic excellence: Graduated top-3 with highest honors in Computer Science (Artificial Intelligence) from Alamein International University on a Full Presidential Scholarship, and co-authored 8+ peer-reviewed papers across IEEE and Springer in biomedical deep learning, physiological signal decoding, and NLP.

Toolbox

What I build with

Languages

PythonC++CJavaSQLJavaScriptTypeScript

AI & Deep Learning

PyTorchTensorFlowKerasCUDAScikit-learnOpenCVPandasNumPyJupyterMLflowWeights & BiasesStreamlitGradioDeep LearningSignal Processing

Speech & Audio AI

HuggingFaceWhisperNVIDIA NeMoNVIDIA MagpieWav2Vec2Meta MMSSpeech AIForced AlignmentAudio AlignmentSpeech SynthesisSpeaker DiarizationDialectal ArabicAcoustic ModelingAudio ProcessingVADCTC SegmentationvLLM

LLMs, RAG & Agents

Llama 3LangChainLangGraphQdrantOllamaCrewAIRAG pipelinesVector embeddingsPrompt EngineeringFew-Shot PromptingLLM-as-a-JudgeMCP (Model Context Protocol)Agentic AIGenesysSession AnalyticsConversation IntelligenceNLPTF-IDFOCR

Cloud & Infrastructure

GCPAWSAzureCloudflareDockerKubernetesLinuxGitGitHub ActionsPyPICI/CDKServeCLI ToolingMLOps

Backend & Databases

FastAPIFlaskPostgreSQLMySQLRedisSQLiteSQLAlchemyCeleryPostman

Frontend & Frameworks

ReactNext.jsTailwind CSSNode.jsHTML5CSS3

Aug 2026 – PresentRemote, Cairo

Fine-tuning Egyptian Arabic TTS models for voice AI and Saeedi dialect ASR transcription models.

  • Fine-tuned Egyptian Arabic TTS models to generate natural, dialect-accurate speech synthesis for voice AI applications.
  • Fine-tuned ASR models for the Saeedi Egyptian dialect, improving transcription accuracy for an underrepresented regional speech pattern.
Speech AISpeech SynthesisDialectal ArabicAudio ProcessingPyTorchHuggingFaceMLOps

Dec 2025 – Aug 2026Remote, Saudi Arabia

Built an end-to-end production AI voice call system (VAD, STT, LLM, TTS) optimized for real-time conversation latency.

  • Built and deployed an end-to-end AI voice call system (VAD, STT, LLM, TTS), optimizing the pipeline to achieve ∼400 - 800ms inference latency for real-time conversations.
  • Collected and curated Saudi dialect audio data, building a full pipeline to clean, segment, and prepare it for fine-tuning multiple open-source TTS and ASR models for high-fidelity, low-latency Saudi Arabic speech synthesis and transcription.
  • Deployed and maintained a large language model in production using GCP for low-latency streaming inference.
PythonFastAPIGCPDockerWhisperSpeech AIAudio ProcessingvLLM

Dec 2025 – Feb 2026Hybrid, Cairo

Processed and analyzed SEC corporate filings using NLP pipelines for large-scale AI disclosure extraction.

  • Processed and analyzed SEC filings using NLP pipelines for large-scale text extraction and preprocessing.
  • Built labeled dataset of AI-related corporate disclosures and engineered semantic features for quantitative analysis.
  • Applied text mining techniques (TF-IDF, keyword extraction, co-occurrence analysis) to identify disclosure themes and patterns.
PythonFastAPINLPTF-IDFPandasNumPyScikit-learn

Jul 2025 – Dec 2025Remote, Cyprus

Developed multimodal RAG pipelines for unstructured documents and natural-language-to-SQL database agents.

  • Built an AI system to extract text from various file formats, normalize output, and persist clean structured data.
  • Developed a multimodal Retrieval-Augmented Generation (RAG) pipeline for PDFs and images using OCR for semantic retrieval and context-aware generation.
  • Designed and deployed text-to-SQL agents to translate natural language queries into executable database queries with automated chart visualisations.
PythonLangChainFastAPIPostgreSQLRAG pipelinesVector embeddingsOCR

Feb 2025 – Apr 2025Remote, Saudi Arabia

Built an Arabic speech recognition engine with real-time error detection for mispronounced words and stuttering.

  • Built an Arabic ASR system with error detection for mispronounced words and stuttering using TTS/STT pipelines.
  • Developed automated monthly analytics reports to monitor speech correction progress.
  • Engineered deep learning models for stuttering detection and submitted research findings to Interspeech.
PythonPyTorchWhisperSpeech AIAudio Processing

Aug 2023 – Jan 2025Cairo, Egypt

Applied deep convolutional neural networks and wavelet transforms to cardiovascular time-series signals and proteomic data.

  • Co-authored ~8 research papers applying AI to medical/physiological data and computer science, presented at international conferences.
  • Applied Continuous Wavelet Transforms to blood-pressure time series to generate scalograms, training a CNN that achieved 94% test accuracy.
  • Used saliency maps to identify critical frequency peaks for disease risk stratification and early prediabetes detection.
PythonPyTorchTensorFlowOpenCVSignal ProcessingDeep Learning

Selected work

Four things worth opening

Each of these has a full write-up — highlights, stack, screenshots and links.

01 / 2026 · System

ASRX

Universal ASR Forced Alignment & Diarization Engine

A model-agnostic speech alignment engine that synchronizes audio with transcripts from any ASR model or LLM down to millisecond precision.

🏆 Published on PyPI · pip install asrx

PythonPyTorchWav2Vec2Meta MMSNVIDIA NeMoForced AlignmentSpeaker DiarizationCTC Segmentation

02 / 2026 · Research

Magpie TTS Arabic

Saudi & Emirati Dialect Speech Synthesis

Dialect-adapted multilingual speech synthesis adapting NVIDIA's Magpie architecture to conversational Gulf Arabic.

🏆 200+ Downloads on Hugging Face

PythonPyTorchNVIDIA MagpieHuggingFaceDialectal ArabicAcoustic ModelingSpeech SynthesisSpeaker Diarization

03 / 2025 · System

Harmonia ML

Zero-Dependency CUDA & PyTorch Stack Auditor

A lightweight, zero-dependency CLI auditor that eliminates CUDA version mismatches and PyTorch ecosystem dependency conflicts before you start training.

🏆 Published on PyPI · pip install harmonia-ml

PythonCUDAPyTorchPyPICLI ToolingMLOpsDocker

04 / 2024 · Product

AIQ MOOC Engine

RAG-Powered Dynamic Question Generation

Parameter-driven RAG architecture that transforms video lecture transcripts into validated, high-quality multiple-choice questions.

🏆 Presented at Int'l Conference on Universities & AI

PythonLlama 3WhisperQdrantFastAPILangChainRAG pipelinesFew-Shot Prompting

The rest of it

Education, awards, writing

Education

  • BSc. in Computer Science (Artificial Intelligence Program)Alamein International University — AIUTop 3 Ranked · Full Presidential Scholarship · Oct 2020 – Jun 2024
  • Software Engineering InternInformation Technology Institute (ITI)Client-server architecture & network communication · Feb 2022 – May 2022

Awards

  • Full Academic ScholarshipPresident El-Sisi Scholarship for Academic Excellence
  • Top 3 Academic HonorAI Science Program, Alamein International University
  • Open Source Community Impact1,000+ Downloads on Hugging Face

Let's buildsomething.

Open for cool builds and interesting problems. The fastest way to reach me is email, I answer all of them.

ahmedkhaled553311@gmail.com
Quick Note
Direct to inbox

Send me a message

Have a project in mind, an opportunity, or a quick question? Drop a note below and I'll get back to you.

Or write directly:
Ahmed Eladl · Cairo, EgyptThanks for scrolling all the way down :)