Halide Yılmaz

Companies

Companies I co-founded

• HatsonTech

  Hatson Teknoloji Ltd. Şti. · August 2026, Ankara

  Software and AI

  We build AI agents, source-citing RAG assistants, language-model fine-tuning, computer vision, data engineering, and web and mobile software for organisations.

  hatsontech.com

• Caseon AI

  Caseon Teknoloji A.Ş. · 2026

  Legal technology

  We are building software for lawyers’ due diligence and contract review work. Every answer is shown with the document paragraph it relies on, and the final check always stays with the lawyer.

  caseon.ai

Today I spend my time on the growth, products and teams of these two companies.

Basics7 min read

What Is Machine Learning? Computers That Learn From Examples

A plain-language guide to machine learning: learning from examples, supervised and unsupervised learning, training versus prediction, and what ML is not.

Read more

• 
  Governance10 min

  The EU AI Act's New Timeline: What the 2026 Omnibus Changed and What Is Actually in Force Today

  The EU AI Act was not delayed; part of it was. What Regulation (EU) 2026/1744 actually changed, why Article 50 transparency has applied since 2 August 2026, and how the 2 December 2027 and 2 August 2028 deadlines now work, traced to primary sources.

• 
  Governance9 min

  High-Risk AI Systems: What an ML Team Actually Has to Build Before December 2027

  The Article 9-15 obligations for high-risk AI systems, translated from legal language into engineering deliverables: a risk register, data governance, Annex IV technical documentation, event logging, an accuracy declaration and a human oversight design.

• 
  Governance9 min

  Labelling AI-Generated Content: Article 50, Machine-Readable Watermarks and C2PA

  Article 50 transparency obligations took effect on 2 August 2026 and were not deferred. What machine-readable marking actually means, hard binding versus soft binding, C2PA content credentials, and why a watermark usually dies at the first re-encode.

• 
  Governance8 min

  ISO/IEC 42001 for Engineers: What an AI Management System Actually Requires — and Why It Is Not AI Act Compliance

  What ISO/IEC 42001 actually requires, how its Annex A controls map onto real ML artefacts, what ISO/IEC 42005 and 42006 added, and why the standard is not harmonised and therefore grants no presumption of conformity under the EU AI Act.

• 
  Governance8 min

  The NIST AI Risk Management Framework in Practice: Turning Govern-Map-Measure-Manage into Engineering Work

  Instead of restating the four functions of the NIST AI RMF, this post turns them into engineering work. Why the MEASURE function is the one most teams leave empty, what the Generative AI Profile added, and why the framework is currently under revision.

• 
  Governance9 min

  AI Impact Assessment: How FRIA, ISO 42005 and a DPIA Differ — and How to Run One

  A fundamental rights impact assessment, an ISO/IEC 42005 impact assessment and a data protection impact assessment are three different documents. Who each one binds, when it is required, when it applies, and how to run them together without doing the work three times.

• 13 Aug 2026Fairness Auditing in Practice: How to Actually Audit a Model — and Why Most Audits Are Theatre8 min
• 13 Aug 2026AI Incident Reporting and Post-Market Monitoring: The 15-Day, 10-Day and 2-Day Clocks8 min
• 13 Aug 2026Data Provenance and Training-Data Licensing: Can You Prove Where Your Dataset Came From?8 min
• 13 Aug 2026Data Contracts for ML: Stopping the Schema Change That Silently Breaks Your Pipeline8 min
• 13 Aug 2026Datasheets for Datasets: The Missing Half of the Model Card8 min
• 13 Aug 2026Annotator Agreement and Evaluator Reliability: Kappa, Alpha, and Validating an LLM Judge10 min

Interests & work

• Large Language Models

  Fine-tuning and evaluation work to adapt language models to Turkish and to specific domains.

• Natural Language Processing

  Classification, information extraction and meaning-based processing of Turkish and multilingual text.

• RAG

  Retrieval architectures that ground answers in the relevant documents and show their sources.

• Semantic Search & Vector DB

  Search systems that consider meaning as well as keywords.

• Data and Model Evaluation

  Data quality, labelling, and evaluating models with the right metrics.

• Legal Software

  Software for legal workflows such as due diligence and contract review.

• MCP & Agent Systems

  Agent workflows that use tools and wait for human approval at important steps.

• System Architecture & MLOps

  Architecture of web and backend systems; training, deploying and monitoring models.

Journey

1. Bachelor’s

   METU · Computer Engineering

   Bachelor’s degree.

2. 2024

   Our first company

   We founded our first company, in AI.

3. January 2025

   Large-scale AI work

   Large-scale model training work with TÜBİTAK, TRUBA and EUMaster4HPC.

4. 2025

   METU Technopark

   Our legal-software project took us into METU Technopark.

5. 2026

   Investment

   Our legal-software project received investment.

6. July 2026

   Project sale

   We sold the legal-software project to another company.

7. August 2026

   HatsonTech

   Our main company; we build software and AI products across fields.

8. 2026

   Caseon AI

   The legal-technology company we founded for contract analysis.

9. Today

   Environment, law and AI

   We build technology products in environment, law, software and AI.