ABDELKERIM DASSI
AI Engineer & Data Scientist
Available for opportunities
ABDELKERIM
DASSI

Building AI systems that ship — RAG pipelines, LLM architecture, and production MLOps at scale. Tunis → World.

Scroll
95%
Classification accuracy
1M+
Records processed
40%
Throughput increase
4×
NVIDIA DLI certified
/ 01 — Work history

Experience

01
AI Engineer
1morething Ventures — Tunisia · Hybrid
  • Architect and deploy scalable ML products across B2B portfolio — MEA & Central Asia startups
  • Define AI strategy, productionize RAG / LLM solutions, mentor engineers on MLOps & model governance
Jan 2026
Present● Live
02
Data Science Instructor
RCH International · TeachCode · GOMYCODE — Remote & Hybrid · Freelance
  • Teaching AI Full Pack across three academies — Python, ML, Deep Learning, Generative AI, Transformers
  • Curricula in English, French and Tunisian dialect — praised for clear, practical, hands-on delivery
Mar 2025
Present● Live
03
Data Scientist
Qualipro® by Imagine Human — Tunisia · Hybrid
  • AI integration into QHSE platform — real-time translation, AI assistant, audit automation agents
  • Full AI delivery with SpringAI, RAG, MCP, Groq — measurable productivity gains
Oct 2025
Feb 2026
04
Data Scientist
D2D Analytics — Canada · Remote
  • 100K+ parliamentary speeches — topic classification pipeline at 95% accuracy
  • Designed & deployed RAG-driven LLM for citizen query answering
Apr 2025
Sep 2025
05
AI Engineer
WEVIOO — Tunisia · Hybrid · Internship
  • Led RAG system research & LLM selection — AWS SageMaker prototyping for NTTDATA
  • Cloud pipelines on S3, EC2, EKS — Docker + GitLab CI/CD
Feb 2024
Dec 2024
06
Data Scientist
Silver Brain AI AG — Zurich, Switzerland · Remote
  • RoBERTa NLP pipeline — 1M+ German legal records, +15% downstream model accuracy
  • Automated processing — 40% throughput increase
Jun 2023
Oct 2023
/ 02 — Selected work

Projects

01
Parliamentary RAG System
RAG · LangChain · AWS · Python
95% accuracy
02
German Legal NLP Pipeline
RoBERTa · HuggingFace · PyTorch
1M+ records
03
Vehicle Pose Recognition API
CNN · FastAPI · Docker · CI/CD
91% precision
04
QHSE AI Platform
SpringAI · RAG · MCP · Groq
Production
05
DocMate — AI Documentation Assistant
RAG · Weaviate · SageMaker · EKS · Docker
Enterprise
/ 03 — Live from GitHub

Activity

@Abdelkerim-Dassi Synced live
Contribution graph — refreshed from GitHub on every visit. View profile
/ 04 — Technologies

Stack

LLMsRAGLangChain PyTorchTensorFlowAWS SageMaker DockerCrewAIKubernetes Fine-tuningFastAPIMLOps NLPComputer VisionRoBERTa LLMsRAGLangChain PyTorchTensorFlowAWS SageMaker DockerCrewAIKubernetes Fine-tuningFastAPIMLOps NLPComputer VisionRoBERTa
AI / LLMs
  • LLMs & Prompt Engineering
  • RAG Systems
  • LangChain / CrewAI
  • Fine-tuning & LoRA
  • HuggingFace
  • MCP · Groq · SpringAI
ML / DL
  • PyTorch / TensorFlow
  • Computer Vision (CNN)
  • NLP / RoBERTa
  • Scikit-learn
  • Transfer Learning
  • Anomaly Detection
Cloud & MLOps
  • AWS SageMaker
  • S3 / EC2 / EKS
  • Docker / Kubernetes
  • GitLab CI / GH Actions
  • FastAPI
  • CI/CD Pipelines
Languages
  • Python
  • Java
  • SQL
  • Bash
/ 05 — Credentials

Certifications

Fundamentals of Deep Learning
NVIDIA DLI
Certified
Building Transformer-Based NLP
NVIDIA DLI
Certified
AI for Anomaly Detection
NVIDIA DLI
Certified
Computer Vision for Industrial Inspection
NVIDIA DLI
Certified
National Engineer Degree — Data Science
ESPRIT School of Engineering · 2024
Degree
Higher Education — Maths & Physics
Tunis Preparatory Engineering Institute · 2021
Degree
/ 06 — From colleagues & mentors

Reviews

"

Abdelkerim has a good and solid understanding of RAG systems and deploying LLMs on AWS, and approached his work with curiosity and a structured mindset. He was proactive in solving challenges and took the time to deepen his understanding when needed.

WB
Wassim Boubaker
Technical Project Manager & Lead Consultant · NTT DATA DACH
Co-advisor
"

Your clear and practical way of teaching complex topics — from Python fundamentals to machine learning, deep learning, Generative AI, and Transformers — made the entire learning journey truly enriching. Your passion, patience, and continuous support greatly strengthened my understanding.

YN
Yasmin Nabli
Embedded Systems & Full Stack JS Engineer
Student
"

Your real-world examples and clear guidance revealed the incredible potential of AI. We truly appreciate your patience as you led us through hands-on projects, deep-learning fundamentals, and encouraged us to explore beyond the basics. You've inspired us to apply these skills in real-world projects.

SR
Sabri Rakrouki
Full-Stack Software Engineer · Microservices
Student
"

J'ai particulièrement apprécié sa grande motivation et sa capacité à apprendre rapidement. Il est une personne autonome et sérieuse. Je suis convaincue qu'il réussira dans sa future carrière.

NA
Nivine Attoué
Data Scientist · R&D Engineer · PhD
Mentor
/ 07 — Essays & articles

Writing

001
RAG in Production: What Nobody Tells You
Beyond basics — chunking strategies, retrieval quality, re-ranking, and the subtle failures that kill accuracy in real deployments. Written from 3 production systems.
RAG LLMs Production
Read
002
I Used Pip for Years. Poetry for Teams. Then Uv Showed Up.
An honest comparison of pip, poetry, and uv from a working AI engineer. Written plainly, in five minutes.
Python Tooling AI Stack
Read
003
From Notebook to AWS: My MLOps Stack
Going from a working model to a production-grade pipeline — SageMaker, EKS, GitLab CI, without the fluff. The exact setup I've used across 4 deployments.
MLOps AWS Docker
Soon
004
Fine-tuning LLMs Without Burning Cash
LoRA, QLoRA, and the real cost-quality tradeoffs. When fine-tuning beats prompting, and when it doesn't — from actual experiments, not benchmarks.
LLMs Fine-tuning LoRA
Soon
Let's build
something
together.
Open to senior AI/ML engineering roles, consulting, and interesting problems.

Based in Tunis, Tunisia.
Available globally — remote friendly.

Response within 24h.