diff --git a/.github/actions/spelling/allow/names.txt b/.github/actions/spelling/allow/names.txt
index 0ffa85c..d7d52c3 100644
--- a/.github/actions/spelling/allow/names.txt
+++ b/.github/actions/spelling/allow/names.txt
@@ -22,6 +22,7 @@ Dmitry
 Duswald
 Elrawy
 Foco
+Galin
 Graenitz
 Guilherme
 Guiraud
@@ -123,6 +124,7 @@ atre
 atreshreyas
 baidyanath
 bellei
+bistrev
 chaitanya
 chaitanyashahare
 chaudhari
@@ -137,6 +139,7 @@ efremale
 elrawy
 fransham
 fsfod
+galin
 gargvaibhav
 garima
 garimasingh
diff --git a/.github/actions/spelling/allow/terms.txt b/.github/actions/spelling/allow/terms.txt
index 12d7045..a580bbf 100644
--- a/.github/actions/spelling/allow/terms.txt
+++ b/.github/actions/spelling/allow/terms.txt
@@ -24,6 +24,7 @@ LLVM
 LULESH
 NVIDIA
 NVMe
+Ohridski
 OMP
 OpenMP
 PTX
@@ -40,6 +41,7 @@ backpropagation
 biodynamo
 bioinformatics
 blogs
+cms
 consteval
 cppyy
 cytokine
diff --git a/_posts/2025-09-07-implementing-AD-in-CMS-Combine.md b/_posts/2025-09-07-implementing-AD-in-CMS-Combine.md
new file mode 100644
index 0000000..bd8a521
--- /dev/null
+++ b/_posts/2025-09-07-implementing-AD-in-CMS-Combine.md
@@ -0,0 +1,75 @@
+---
+title: "Implementing AD in CMS Combine"
+layout: post
+excerpt: "A  CERN Summer Student 2025 project aiming at the integration of 
+automatic differentiation (AD) into the CMS Combine tool to accelerate statistical
+inference by leveraging RooFit's AD support and LLVM-based gradient generation."
+sitemap: false
+author: Galin Bistrev
+permalink: blogs/2025_galin_bistrev_introduction_blog/
+banner_image: /images/blog/gsoc-banner.png
+date: 2025-07-05
+tags: cern cms root combine c++ rooFit automatic-differentiation 
+---
+
+### Introduction
+Greetings! I’m Galin Bistrev, a fourth-year student specializing in Nuclear and 
+Particle Physics at the University of Sofia "St. Kliment Ohridski".
+As part of the CERN Summer Student Programme 2025, I’m working on a project to 
+integrate automatic differentiation (AD) 
+into the CMS Combine tool.
+
+### Project description
+
+This project focuses on implementing automatic differentiation (AD) into the CMS 
+Combine tool, which the primary statistical analysis framework used by the CMS experiment
+at CERN. Combine is built on top of RooFit, which recently introduced AD 
+technology  to support  minimization methods.
+By generating computationally efficient gradients through AD, RooFit enables significant
+performance gains. RooFit’s implementation of AD works by converting internal 
+likelihood representations into standalone C++ code, from which gradient code is 
+produced. This approach not only accelerates fitting but also improves 
+the portability and shareability of likelihood models, making them accessible
+even to users without deep knowledge of RooFit or Combine internals.
+
+### Brief description of the CMS Combine engine
+Combine is a statistical analysis tool designed to compare a model of expected 
+observations with real data. It’s commonly used for tasks such as discovering 
+new particles or processes, setting limits on potential new physics, and measuring
+physical quantities like cross sections.While Combine was developed with High 
+Energy Physics (HEP) applications in mind, it contains no built-in physics 
+knowledge,thus it remains completely general and independent of the interpretation
+of any specific analysis. This flexibility allows it to be used across a wide range 
+of statistical problems
+
+### Project goals
+The main goals of this project are:
+
+- Support external users in working with Combine-generated models without needing
+to dive into RooFit or Combine internals.
+
+- Optimize performance in high-complexity workflows within the Combine framework.
+
+
+### Implementation strategy 
+
+- Refactor Combine to use standard RooFit primitives where possible, and add AD support to custom components.
+
+- Integrate RooFit's AD- generated gradients into Combines's likelihood  
+evaluation and minimization workflows to improve performance and enable external 
+gradient use.
+
+
+- Developing benchmarks to quantitatively demonstrate the performance improvements AD brings to Combine overflows.
+
+## Conclusion 
+
+By integrating automatic differentiation into the CMS Combine tool, we aim to make
+statistical inference not only faster but also more accessible to the broader 
+community. I’m excited to contribute to this effort during my time at CERN and 
+look forward to sharing progress, insights, and benchmarks as the project evolves.
+
+### Related Links
+- [CMS Combine GitHub page]https://cms-analysis.github.io/HiggsAnalysis-CombinedLimit/latest/
+- [ROOT official repository]https://github.com/root-project/root
+- [My GitHub profile]https://github.com/GalinBistrev2 
diff --git a/images/blog/banner-cern.jpg b/images/blog/banner-cern.jpg
new file mode 100644
index 0000000..1b4145e
Binary files /dev/null and b/images/blog/banner-cern.jpg differ