Bardia Shahbaz
Senior Function & Concept Engineer · Product Owner
Senior engineer with 15 years of experience in function development and systems architecture for highly automated driving. Expert in safety-critical systems (Vehicle Motion Management / ESC) with deep expertise in ISO 26262, ASPICE, and MBSE (Matlab/Simulink). Experienced in virtual validation (vECUs), SiL/HiL methods, and agile requirements management in a large OEM environment.
Experience
03.2025 – present
Ingolstadt / Hybrid
Senior Function & Concept Engineer
CARIAD SE
- Project lead and successful series launch of safety-critical modules (hold & standstill management, FailSafeStop) within a global platform software.
- Concept and design of the logical and technical system architecture for highly automated driving functions.
- End-to-end process ownership (requirements-to-release) to safeguard functional requirements and safety standards.
- Primary point of contact for change and stakeholder management to align technical requirements.
- Systematic root-cause analysis and lasting problem resolution to safeguard software quality and system stability.
07.2022 – 02.2025
Ingolstadt / Hybrid
Senior Function & Concept Engineer
CARIAD SE
- Strategic introduction of continuous, development-integrated SiL testing to accelerate release cycles through early defect detection.
- Continued series support for deceleration management until the successor system was in the field, including analysis and resolution of customer complaints.
- End-to-end validation of complex driving functions using vECUs and vehicle tests to reduce hardware dependencies.
- Technical integration of ~70 sub-modules into the overall VMM software, ensuring runtime on the platform.
- Technical alignment of requirements between module developers and the overarching architecture.
- Responsible for build management, configuration, and deployment of software releases.
01.2017 – 06.2022
Ingolstadt
Function Owner, Motion Management, with continuing series support for Deceleration Management
AUDI AG
- Development of the logical and technical function architecture for Vehicle Motion Management (VMM) in automated driving.
- Series support for deceleration management continued in parallel as the system scope grew to motion management.
- Design and build-out of a simulation-based development environment in Matlab/Simulink for VMM.
- Development of automated evaluation and validation tools for measurement data from testing, plus prototype implementation and validation of driving functions in vehicle testing.
10.2014 – 12.2016
Ingolstadt
Calibration Owner, later Function Owner, Deceleration Management
AUDI AG
- Calibration, commissioning, and validation of deceleration management in vehicle testing across multiple vehicle platforms.
- Grew out of calibration ownership into the function design itself, through to function ownership and series release of the module.
- Directed requirements, change, and defect processes in close coordination with system suppliers and specialist departments.
- Ensuring homologation readiness and release quality in strict compliance with ISO 26262 and market-specific quality standards.
08.2011 – 09.2014
Ingolstadt
Function Integration / Calibration, Deceleration Management
EFS GmbH / Wotech GmbH (for AUDI AG)
- Commissioning, calibration, validation, and series release of the deceleration-management module within the electronic stability control (ESC).
- Defect and change management to ensure stable function development.
09.2009 – 05.2011
Oberpfaffenhofen
Student Assistant
DLR – Institute of Robotics and Mechatronics
- Diploma thesis: development of a haptic feedback device for the human foot.
- Internship: concept study for a haptic feedback device on the foot.
08.2007 – 08.2009
Dresden
Student Assistant
Fraunhofer IZFP-D
- Behavioral simulation of piezo materials for non-destructive testing methods.
Applied AI
2023 – present
Remote / Germany
Independent AI Architect & Engineer
Independent
Document pipeline
- Stage 1: LLM classification of incoming mail (10–15 letters a week) by addressee and subject, with deterministic routing into filing, tasks, or the financial stage.
- Stage 2: extraction of ~15 financial documents a day into typed schemas behind a multi-provider fallback (Mistral OCR, Landing AI, Pydantic).
- Provider selection across four generations (Nanonets, GPT-4o, Landing AI, Mistral OCR), assessed at line-item level and made comparable by an adapter onto a single internal contract.
- Unattended operation via a schema- and prompt-versioned extraction cache, content-fingerprint dedup, and quarantine for validation failures.
Content pipeline
- Copywriter and critic as separate roles, bound to hard brand and identity anchors against invented products and health claims.
- The most recent critic findings are fed back from a deduplicated store into the next prompt.
- Image generation onto a deterministically rendered layout: the model may only fill placeholder anchors with icons, without altering text or layout.
- Daily posting since July 2026 with weekly batched approval, and no publishing credentials held by the generator.
Operations system
- Scheduling, inventory, and supplier ordering (React Native, Expo, Supabase), in daily use since 2025.
- Price comparison across six B2B suppliers with fuzzy-matched product normalization over three differently shaped sources.
Education & Certifications
09.2005 – 05.2011
Dresden
Mechatronics (Dipl.-Ing.)
Technische Universität Dresden
- Focus areas: robotics, hydraulics / pneumatics.
2026
Udacity
AI / ML Nanodegrees
Udacity
- AI Programming with Python; ML DevOps Engineer.
2017
Coursera
Deep Learning Specialization
DeepLearning.AI
Skills
AI & Machine Learning
Safety-Critical Systems & Architecture
Programming
Tools & Platforms
Languages
- German
- C2
- English
- C1
- Spanish
- A1
- Persian
- Native