Baris Sozudogru

Curious Engineer. TUM. Munich.

Now

I manage the infrastructure and architecture at taico, where we run a GCP platform serving major European enterprises in media analytics, retail, and streaming. I also consult for Telefonica (O2) on 5G edge computing and computer vision for autonomous systems.

Day to day, I work across the full stack: cloud infrastructure on GCP, API development, data pipelines, databases, and agentic AI workflows. My academic background is in robotics, computer vision, and autonomous systems (TUM).

In my own time I build things: an artificial life simulation in Rust/WebAssembly, multi-agent systems with Google ADK, MCP servers, and context engines for AI coding assistants. Details on each project below.

Work

Engineering Manager, taico GmbH 2024 – present

Leading architecture, infrastructure, and security for a GCP-based analytics platform serving 5 enterprise clients in media analytics, retail, and streaming.

Own technical roadmap and architecture decisions across the platform (Python/Flask, React, BigQuery, Cloud Run)
Dashboard load times reduced from 10s to under 1s through query optimization and Redis caching
BigQuery scan costs cut by 25–100x via table partitioning, clustering, and materialized views
Led security incident response and remediation; executed platform-wide security audit covering firewall rules, IAM policies, and access controls
Established engineering security practices: audit logging across 8 GCP services, PII hashing, least-privilege IAM, and vulnerability remediation pipeline
Built internal monitoring system (Pulse Sight) with Prometheus metrics, alerting, and capacity forecasting for infrastructure health and cost tracking

Technical Consultant (Part-time), Telefonica (O2) Germany 2025 – present

Part-time technical advisory on 5G, edge computing, and computer vision at the Technology Strategy & Innovation department.

Designed edge computing reference architectures for sub-millisecond object detection across industrial automation, cobots, drones, and AMR devices
Developed use case demonstrations for stakeholders ranging from C-level to department heads and B2B partners
Technical assessments and reporting on intralogistics automation and edge-based platform management systems

Founding Engineer, taico GmbH 2021 – 2024

First engineer. Built the platform from scratch: data pipelines, APIs, cloud infrastructure. Grew the company from idea to funded product.

Designed and built the full data platform in Python: 22 automated Airflow pipelines, 117 Flask API endpoints, 1,360+ BigQuery tables with dbt transformations
Set up GCP infrastructure from zero (Cloud Run, Cloud SQL, Pub/Sub, Docker, GitHub Actions CI/CD) and deployed with Terraform
Secured seed funding and hired the initial engineering team
Sole backend engineer for the first two years, scaling the platform from first customer to five enterprise accounts

Education

M.Sc. Robotics, Cognition and Intelligence 2021 – 2024

Technical University of Munich.

B.Sc. Electrical and Computer Engineering 2017 – 2021

Technical University of Munich.

Research

Private 5G for Real-Time Object Detection 2024

Master's thesis with Telefonica. Compared 5G and Wi-Fi for autonomous systems on an industrial forklift. Presented to Telefonica supervisory board.

Motion Prediction for Autonomous Driving 2023

Scene-aware and social-aware neural networks for vehicle behavior prediction. Chair of Cyber Physical Systems, TUM.

Radiology Report Generation with LLMs 2023

Large-scale ground truth dataset for automated structured radiology reports. Chair of Computer Aided Medical Procedures, TUM.

Point Cloud Registration with Graph Neural Networks 2022

Localization for autonomous driving via GNN-based point cloud registration. Chair of Robotics, AI and Real-time Systems, TUM.

CNNs and Extreme Learning Machines on Mobile Devices 2021

Bachelor's thesis. Deep learning for image classification on Raspberry Pi. Chair of Data Processing, TUM.

Projects

Petri Labs

Artificial life simulation where neural network organisms evolve, hunt, and speciate in real-time in the browser. Rust/WebAssembly with TypeScript frontend.

Cortex AI

Context engine for macOS that watches screen activity, reads calendar, and assembles relevant context for AI coding assistants via MCP.

ADK Agent Playground

Multi-agent systems built with Google Agent Development Kit. Deep research, code review, structured debate, market analysis, summarization, and iterative writing. Python, model-agnostic via OpenRouter.

Deep Research Digest

MCP server for federated academic search across arXiv, PubMed, and Semantic Scholar.