Anass Allak

SalamDarija

I'm Anass, a

AI Solution Architect

I take messy real-world data — ultrasound video, bank records, spoken Darija — build models around it, evaluate them honestly, and ship them to production.

Anass Allak

Anass Allak

Speaking on Darija speech recognition

Now

Full Stack Data Scientist at Deepecho

Fetal ultrasound AI, from raw video to production.

Based in

Casablanca, Morocco

00:00:00

Overfitting playground

Click to add data, click a dot to remove it. Then crank the degree.

R² = 0.955

A straight line. Simple, honest, maybe too simple.

0+

Years in industry

0+

Projects shipped

0

Research papers

0+

Technologies

Explore

Jump to any project or section from the keyboard. And psst — try typing salam.

Real data is messy. I build models that survive it — for doctors, banks & insurers and people who speak Darija.

I'm a Full Stack Data Scientist working across machine learning, data engineering and applied AI, and a PhD candidate in Machine Learning at the National Institute of Statistics and Applied Economics (INSEA). My work spans clinical fetal ultrasound AI, speech recognition and language models for Moroccan Darija, fraud, credit-risk and churn models for banks and insurers, and the data pipelines and serving infrastructure that keep it all running.

  1. Data
  2. ML
  3. Deep learning
  4. Vision & NLP
  5. Production ML
  6. Cloud & MLOps
  7. AI research
  • 🏥 Medical AI
  • 👁️ Computer vision
  • 🗣️ Speech AI
  • 🇲🇦 Darija NLP
  • 🏦 Banking
  • 🛡️ Insurance
  • 📈 Economic data
  • 🌱 ESG
  • 🔐 Medical privacy
  • ☁️ Cloud & MLOps
  • 🔬 Research

How I work

Four pillars

Research

From Darija NLP to fetal ultrasound — peer-reviewed, cited, published.

Ship

Models in production, not just notebooks. FastAPI + Triton + ONNX on GCP.

Pipelines

Video to dataset to model to serving. Owned end-to-end.

Rigor

Site-generalization eval, failure analysis, honest metrics.

Any modality

Any modality. Data is data.

I architect ML systems across every data modality: structured tabular, text and language, voice and audio, image, video. The modeling, data engineering, and MLOps patterns are the same — the signal just lives in a different shape.

Tabular

fraud, credit risk, churn

Language

Moroccan Darija ASR, Arabic NLP, LLM pipelines

Voice / Audio

ASR, speaker ID, acoustic diagnostics

Image

fetal ultrasound, DICOM, medical CV

Video

ultrasound sweeps, temporal CV, action recognition

By the numbers

Research footprint

0+

Citations

Google Scholar, across venues

0

Peer-reviewed papers

Including NeurIPS workshops

0

h-index

Scholar, lifetime

Selected work

Everything is connected

Medical AI Speech & Darija NLP Finance & Insurance Data & MLOps Vision & Signals Side projects

Hover to trace links · click a project · drag nodes around

Experience

The path so far

  1. Full Stack Data Scientist

    2024 — Present

    Deepecho

    Building fetal ultrasound AI end to end — from raw video and annotation to clinical models, evaluation and production serving.

    • Fetal anatomy segmentation and sweep classification
    • Presentation, placenta, amniotic fluid, biometry and heart-rate models
    • Automated fetal ultrasound reports
    • Site-generalization evaluation (LODO, AUROC) and failure analysis
    • Video-to-dataset pipelines, CVAT annotation and DICOM anonymization
    • Production serving with FastAPI, Triton and ONNX on Google Cloud
  2. PhD Candidate, Machine Learning

    2019 — Present

    INSEA — National Institute of Statistics and Applied Economics

    Doctoral research on NLP and speech recognition for Arabic dialects.

    • First author at the NeurIPS 2021 Data-Centric AI workshop
    • Open-source voice datasets for underrepresented languages
    • Transformer models for Arabic dialects
    • International research collaborations
  3. Data Scientist / Engineer

    2024

    Aiox-labs

    Led AI and data projects for major Moroccan banks, insurers and infrastructure operators.

    • Inflation index pipelines for Bank Al-Maghrib
    • Fraud detection (WAFER) and a Darija voice bot for Attijariwafa Bank
    • Missed-payment prediction for Société Générale
    • Customer churn prediction for Wafa Assurance
    • ESG intelligence for ToumAI and canal inspection for Lydec
    • Speech recognition and BERT models for Moroccan Arabic
  4. Machine Learning Intern

    2021

    Aiox-labs

    Voice recognition and generation for the Moroccan Arabic dialect.

    • Built the DVoice speech recognition project
    • Voice generation models for dialectal Arabic
    • Contributed to open-source speech datasets
  5. Research Intern

    2018

    ENSA Tétouan — Telecommunications Laboratory

    Final-year project: an intelligent OFDM radio transmission simulation platform.

    • OFDM simulation platform in LabVIEW
    • Signal processing for radio communications

Research

Publications

Toolbox

Skills & technologies

Filter by domain to see what powers each kind of project.

  • Python
  • TypeScript
  • JavaScript
  • R
  • C
  • Java
  • SQL
  • MATLAB
  • Bash
  • PyTorch
  • PyTorch Lightning
  • TensorFlow
  • Keras
  • scikit-learn
  • Hugging Face
  • Pandas
  • NumPy
  • SciPy
  • OpenCV
  • Jupyter
  • spaCy
  • Whisper
  • BERT
  • Wav2Vec2
  • SpeechBrain
  • DarijaBERT
  • AraBERT
  • DICOM
  • PyDICOM
  • CVAT
  • MONAI
  • nnU-Net
  • 3D Slicer
  • SimpleITK
  • Apache Spark
  • Apache Beam
  • Apache Airflow
  • Hadoop
  • Kafka
  • PostgreSQL
  • Elasticsearch
  • BigQuery
  • Databricks
  • dbt
  • Redis
  • FastAPI
  • Flask
  • Triton
  • ONNX
  • Docker
  • Kubernetes
  • Helm
  • GitHub Actions
  • Jenkins
  • GitLab CI
  • Nginx
  • Google Cloud
  • Azure
  • Cloudflare
  • Vercel
  • Git
  • GitHub
  • Linux
  • Ubuntu
  • VS Code
  • Vim
  • Next.js
  • React
  • Tailwind CSS
  • HTML5
  • CSS3
  • Framer Motion

Teaching · Professor

Courses I teach

As a professor, I design and teach these courses — taking students from fundamentals to the tools I use in production.

Data Analysis

Practical analysis workflows: exploratory analysis, statistical inference, visualization and end-to-end analytical pipelines.

Pandas & NumPyStatistical analysisData visualizationPower BIEDA

Natural Language Processing

From NLP fundamentals to transformers, LLM fine-tuning and low-resource language processing — with a focus on Arabic dialects.

TransformersBERT & LLMsText classificationASRHugging Face

Big Data Engineering

Distributed processing, cloud data pipelines and scalable infrastructure for large datasets in production.

Apache SparkApache BeamHadoopGCP DataflowETL pipelines

Live from GitHub & Hugging Face

Open source

Contact

Let's build something

Have a data problem, a research idea or a role in mind? Send a message and I'll get back to you.

[email protected]

Casablanca, Morocco