class YasserElHaddar:
role = "Tech Lead · Senior MLOps Engineer"
company = "Mindtrace Limited, Manchester 🇬🇧"
focus = ["Computer Vision", "MLOps", "LLM Engineering", "Production ML"]
passion = "Transforming raw data into intelligent, production-grade AI systems"
building = ["ShopSense-AI", "CacheFuse", "SkewSentry", "MCP-DS-Toolkit-Server"]
writing_at = "medium.com/@yasserelhaddar"I lead ML operations for asset inspection and manufacturing defect detection at scale. From architecting serverless GCP pipelines to fine-tuning LLMs with QLoRA — I ship AI that actually works in production.
🛍️ ShopSense-AIMicroservices AI shopping platform with fine-tuned Mistral-7B (QLoRA), semantic search via Qdrant, and real-time product discovery across Amazon, Best Buy & Walmart. Sub-50ms responses for 100+ products per query.
|
Production LLM caching framework achieving 60–90% API cost reduction and 100× latency improvement (3ms vs 2–5s). Drop-in decorator system, multi-backend (SQLite/Redis), privacy-compliant.
|
🛡️ SkewSentryAutomated ML feature validation preventing training/serving skew — addressing 70% of production ML failures. Pre-deployment detection with CI/CD integration and HTML/JSON reporting.
|
Standalone MCP server giving AI assistants (Claude Desktop, Cursor) full data science superpowers. 29 specialized tools for preprocessing, training 14+ algorithms, and experiment tracking.
|
|
AI newsletter system curating personalized content from GitHub, Reddit, HN & web. GPT-4 scoring + LangGraph orchestration — 20 curated articles in 45–60 seconds.
|
Enterprise ML platform for manufacturing defect detection. YOLO/DINOv2 achieving 92% accuracy, point cloud segmentation with 50% faster processing and 60% cost reduction.
|
AI / ML
Backend & APIs
Cloud & Data
LLM & Frameworks
| Period | Role | Company |
|---|---|---|
| 2022 – Present | Tech Lead · Senior MLOps Engineer | Mindtrace Limited 🇬🇧 |
| 2021 – 2022 | Data Scientist | Vitesco Technologies 🇩🇪 |
| 2021 | Research Assistant (M.Sc. Thesis) | Technology Campus Plattling 🇩🇪 |
- 🧠 AlexNet: The Deep Learning Revolution That Sparked The Boom
- 🔷 LeNet-5: The Pioneering CNN That Started It All
- 🔍 HOG: The Foundation of Early Object Detection
- 🌐 4M: The AI Framework Revolutionising Multimodal Learning
- 📦 uv: Redefining Python Environment & Dependency Management
