Aydin Aghaliyev https://aydin.aghaliyev.com/ Recent content on Aydin Aghaliyev Hugo -- gohugo.io en-US aydin.aghaliyev@gmail.com (Aydin Aghaliyev) aydin.aghaliyev@gmail.com (Aydin Aghaliyev) Aydin Aghaliyev Mon, 01 Jan 2024 08:00:00 +0000 About https://aydin.aghaliyev.com/about/ Mon, 01 Jan 2024 08:00:00 +0000aydin.aghaliyev@gmail.com (Aydin Aghaliyev) https://aydin.aghaliyev.com/about/ <h2 id="about-me">About Me</h2> <p>I&rsquo;m <strong>Aydin Aghaliyev</strong>, a Data Scientist &amp; AI Engineer at Philips, based in Rotterdam, The Netherlands. I build agentic AI systems for root-cause analysis and contract reasoning, with experience in knowledge graphs, MCP, NLP, and cloud infrastructure.</p> <h2 id="experience">Experience</h2> <h3 id="philips--amsterdam-netherlands">Philips — Amsterdam, Netherlands</h3> <p><strong>Data Scientist &amp; AI Engineer</strong> — Full-time | Sep 2025 – Present</p> <ul> <li>Architected and built an end-to-end root-cause analysis (RCA) pipeline that ingests incident data from ServiceNow, parses log archives into RDF knowledge graphs stored in Amazon Neptune, and exposes 50+ analysis tools to an LLM agent via MCP (Model Context Protocol)</li> <li>Designed a graph-based investigation algorithm with multi-factor severity scoring and pivot logic, enabling the agent to autonomously narrow investigation scope across thousands of log lines per incident</li> <li>Built a SPARQL pre-execution static analyzer to prevent injection attacks and catch query errors before reaching the database</li> <li>Deployed the MCP server on AWS Bedrock AgentCore; provisioned the entire pipeline infrastructure with Terraform (Lambda, SQS, EventBridge, S3, Neptune), with GitHub Actions CI/CD and OIDC-federated authentication</li> <li>Created an agentic pipeline in Databricks (PySpark) for clause extraction and reasoning over medical service contracts using LLMs</li> </ul> <h3 id="philips--eindhoven-netherlands">Philips — Eindhoven, Netherlands</h3> <p><strong>Generative AI Engineer</strong> — Internship | Apr 2025 – Sep 2025</p> About Me

I’m Aydin Aghaliyev, a Data Scientist & AI Engineer at Philips, based in Rotterdam, The Netherlands. I build agentic AI systems for root-cause analysis and contract reasoning, with experience in knowledge graphs, MCP, NLP, and cloud infrastructure.

Experience

Philips — Amsterdam, Netherlands

Data Scientist & AI Engineer — Full-time | Sep 2025 – Present

  • Architected and built an end-to-end root-cause analysis (RCA) pipeline that ingests incident data from ServiceNow, parses log archives into RDF knowledge graphs stored in Amazon Neptune, and exposes 50+ analysis tools to an LLM agent via MCP (Model Context Protocol)
  • Designed a graph-based investigation algorithm with multi-factor severity scoring and pivot logic, enabling the agent to autonomously narrow investigation scope across thousands of log lines per incident
  • Built a SPARQL pre-execution static analyzer to prevent injection attacks and catch query errors before reaching the database
  • Deployed the MCP server on AWS Bedrock AgentCore; provisioned the entire pipeline infrastructure with Terraform (Lambda, SQS, EventBridge, S3, Neptune), with GitHub Actions CI/CD and OIDC-federated authentication
  • Created an agentic pipeline in Databricks (PySpark) for clause extraction and reasoning over medical service contracts using LLMs

Philips — Eindhoven, Netherlands

Generative AI Engineer — Internship | Apr 2025 – Sep 2025

  • Built a vector-based full-text search system using Gemini embeddings, ChromaDB, FastAPI backend and Vue.js frontend; deployed with key-based authentication on Azure
  • Built a GPT-4o agentic RAG chatbot for in-company use with a full frontend, authentication system, and dynamic memory of previous conversations, deployed on Azure

Education

Erasmus University Rotterdam — MSc Data Science | Sep 2024 – Aug 2025 Coursework: Machine learning, NLP, LLMs, Agentic AI, Web scraping, Python, R

Erasmus University Rotterdam — BSc International Economics and Business Economics (IBEB) | Sep 2021 – Aug 2024 Coursework: Investments, Derivatives, Statistics, Stochastic Calculus, Econometrics, Time Series Analysis, C++, R

Skills

  • Programming languages: Python, TypeScript/JavaScript, C++, R, SQL
  • AI & ML: TensorFlow, PyTorch, LangChain, NumPy, Pandas, Pydantic, RDFLib, Agentic AI, RAG, MCP, Knowledge Graphs, SPARQL, LLM integration (AWS Bedrock, GPT-4o, Gemini)
  • Backend & Data: FastAPI, FastMCP, ChromaDB, Node.js, Vue.js, Spark, Databricks, Power BI, QlikSense, Excel
  • Cloud & Infrastructure: AWS (Lambda, SQS, S3, Neptune, Bedrock AgentCore, EventBridge), Azure, Docker, Terraform, CI/CD (GitHub Actions), ServiceNow REST API, Microsoft Office JS API

Certifications

  • GRE — 160V / 170Q (ETS, Oct 2023)
  • IELTS — 8.0 (Mar 2020)

Languages

Azerbaijani (native) · English (fluent) · Russian (fluent) · Turkish (fluent)

Contact

]]>