Serdar Gündoğdu
Let's talk

AI practitioner · Industrial systems

Industrial intelligence,
built for the real world.

I'm Serdar Gündoğdu—an engineer and AI practitioner who turns complex industrial data into usable products, dependable machine learning systems, and faster decisions.

Follow my journey GitHub LinkedIn

Career journey

One path, many engineering lenses.

I began my journey with the fundamentals of mechanics. Industrial engineering taught me to see systems through a multidisciplinary lens, while a master's degree in materials and manufacturing strengthened my research-led approach. As a production engineer and manager, I worked with lean flows, quality, Six Sigma, and leadership. I later combined that industrial foundation with data science, software, and cloud engineering to build AI products. Today I deliver projects with Claude and Codex as engineering partners while developing my skills in GPU kernel engineering.

  1. 09

    Now · Learning

    Kernel Engineer — studying

    Developing the next layer of understanding through GPU architecture and kernel engineering fundamentals.

    • GPU architecture
    • Kernel engineering
    • In progress
  2. 08

    AI-native delivery

    AI Practitioner

    Uses Claude and Codex as engineering partners to research, build, review, and ship numerous projects while keeping decisions and accountability human-led.

    • Claude
    • Codex
    • AI delivery
  3. 07

    Product engineering

    Full-Stack AI Engineer

    Connected models to complete products through frontend, backend, APIs, deployment, and cloud engineering.

    • Frontend
    • Backend
    • Cloud
  4. 06

    Data

    Data Scientist

    Moved from process evidence to models, algorithms, data pipelines, and analytical software.

    • Machine learning
    • Python
    • C
    • SQL
  5. 05

    Leadership

    Production Manager

    Led teams and processes with a focus on quality, Six Sigma, repeatability, and continuous improvement.

    • Leadership
    • Six Sigma
    • Quality
  6. 04

    Operations

    Production Engineer

    Translated engineering discipline into stable production flow, value stream mapping, and Lean improvement.

    • Value stream mapping
    • Lean
    • Production flow
  7. 03

    Graduate study

    M.Sc. in Materials and Manufacturing

    Strengthened an evidence-driven approach through materials, manufacturing, and academic research.

    • Academic approach
    • Research
  8. 02

    Education

    Industrial Engineering

    Learned to connect people, processes, technology, and decisions as one operating system.

    • Multidisciplinary
    • Systems thinking
  9. 01

    Education

    Mechanical Engineering

    Built a foundation in mechanics, physical systems, and engineering problem solving.

    • Mechanics
    • Engineering fundamentals

From physical systems

to governed intelligence

to decisions people can trust

Selected work

Products, systems, and tools that move beyond the demo.

A selection of industrial platforms and open-source tools shaped around real workflows—not isolated models.

01Industrial AI platform

SWAPP

An AI-first industrial data workbench that brings signal exploration, trends, statistics, PPM views, governed APIs, and natural-language workflows into one product.

  • Industrial data
  • AI workflows
  • Edge to cloud
Visit SWAPP
02Open-source Python package

PIPolars

A high-throughput bridge from the PI System to Polars DataFrames—designed for bulk extraction, lazy evaluation, and practical caching in industrial analytics workflows.

  • Python
  • Polars
  • PI System
View on GitHub PyPI
03Adaptive learning system

AI Practitioner Dev OS

A project-based operating system for becoming a stronger AI practitioner, with a 12-month adaptive curriculum and agents for planning, building, reviewing, and evaluation.

  • AI agents
  • Learning design
  • Open source
Explore the template
04Sovereign AI R&D initiative

Project PULSAR

An independent R&D initiative for running strong open-weight language models securely and sustainably inside organizational trust boundaries.

  • Open-weight models
  • Local control
  • Trusted AI
Explore PULSAR

How I work

Engineering discipline. AI-native pace.

I combine industrial context, statistical thinking, and modern AI development workflows to close the gap between an idea and a system people can actually use.

01

Frame the decision

Start with the operator, engineer, or business decision—not the model. Define the signal, constraint, and useful action.

02

Build the full loop

Connect data, analysis, APIs, interface, and feedback so the intelligence survives outside a notebook.

03

Make it operable

Design for traceability, human review, deployment constraints, and the realities of industrial environments.

Predictive analyticsTime seriesIndustrial data platformsApplied machine learningAI-assisted software delivery

Open lab

Tools for learning by doing.

Packages, platforms, and interactive experiments that make difficult concepts easier to explore.

01

.NET package

PIWebAPI

A REST layer for PI Points, AF hierarchies, StreamSets, and Event Frames.

 02

Industrial platform

SCADA Nerve

A focused platform for SCADA monitoring, automation, and operational visibility.

 03

Learning platform

Industry-Learn

Applied machine-learning patterns and workflows grounded in industrial problems.

 04

Interactive lab

Scikit-Play

A hands-on playground for understanding and comparing classical ML algorithms.

 05

Time-series lab

Aeon-Play

Interactive experiments for time-series analysis and forecasting with Aeon.

 06

Deep-learning lab

PyTorch-Play

A visual environment for experimenting with neural networks and model behavior.

 07

Learning platform

DSML101

Clear, practical resources for building a foundation in data science and ML.

About

I sit at the intersection of engineering, data, and product delivery.

My background spans statistics, mathematics, mechanical and industrial engineering. That mix helps me see AI as a complete operating system: the data foundation, the model, the interface, the deployment path, and the people making the final decision.

I use AI coding partners such as Codex, Claude Code, and Gemini as part of a disciplined development workflow—accelerating research and implementation while keeping architecture, validation, and accountability human-led.

Read my writing Start a conversation

Have an industrial AI problem worth solving?

Let's build something useful.