David Wright

Senior Software Engineer | Applied ML & ML Infrastructure | High-Performance Systems

david.wright@bluewin.ch · Zürich, Switzerland
linkedin.com/in/david-wright-53076b95 · gitlab.com/david_wright · github.com/dav1d-wright

Senior software engineer and team lead in applied ML and ML infrastructure, with 10+ years building production computer vision, real-time camera, sensor and safety-critical systems. Develops models for liveness, fraud detection and document understanding, and the inference infrastructure that serves them at 10M+ verifications per year. Built streaming pipelines for XR devices at ~1GB/s. Experienced at Meta, startups and global companies.

Professional Experience

Team Lead | PXL Vision AG | Zürich, CHJun 2026 - Present

• Led a team of 3 engineers owning ML inference infrastructure and ML R&D for video injection defence.
• Shipped a proof of concept for a hardened capture platform with a substantially reduced attack surface, and advocated for it, securing company-wide investment in the platform.
• Developed a new liveness detection model using randomised colour illumination to prevent video injection.
• Developed a video manipulation detection model that identifies editing and encoding artefacts, achieving <15% attack error rate (APCER) at <1% bona fide rejection rate (BPCER).
• Led a model performance analysis of CLIP-based deepfake detection across capture sources, traced the root cause to capture and encoding settings, and improved the pipeline, reducing BPCER by up to 10 pp.

Senior Software Engineer | Meta Reality Labs | Zürich, CHJan 2025 - May 2026

• Co-owned a gstreamer-based streaming and recording framework for sensor, camera, and device state (e.g. SLAM) data at ~1GB/s, and solutions built on it.
• Owned a sensor metrics validation tool, defining new metrics and performing online statistical validation (t‑digest) of streaming metadata (drops, jitter, synchronisation) across XR devices.
• Led a platformisation initiative for sensor and camera validation, resulting in a 95% reduction of developer effort for onboarding new XR devices.
• Technical Lead on always-on data collection tool for XR devices. Designed mmio/mmap buffering strategy achieving 300MB/s sustained writes without noticeable user impact.

Team Lead | PXL Vision AG | Zürich, CHJun 2023 - Dec 2024

• Owned the ML inference infrastructure and backend services powering the CV/ML stack, serving 10M+ verifications per year.
• Led a team of 3 engineers owning the platform of the company’s distributed Identity Verification system across SaaS, on-premises, and mobile deployments.
• Designed and led passive injection attack detection using a dual-pipeline architecture (UX-optimised vs. security-optimised) to defend against video replay attacks while maintaining conversion.
• Led upgrades to video frame extraction (quality-based algorithm, +60% data yield) and snippet extraction (text-boundary detection, up to +20 pp OCR acceptance rate).

• Tuned field detection using Optuna hyperparameter optimisation, achieving up to +10 pp detection rate.

Technical Lead | PXL Vision AG | Zürich, CHDec 2022 - May 2023

• Designed embedding-based document identification using nearest-neighbour search, achieving >90% recall.
• Designed gRPC and WebAssembly SDKs for client-side document detection, cutting backend calls by ~50%. Adopted company-wide and by external customers.
• Implemented a nonlinear optimisation backend (Ceres) for document unwarping.

Software Engineer | PXL Vision AG | Zürich, CHOct 2021 - Nov 2022

• Designed an in-memory index and query API unifying document fields from multiple pipelines.
• Designed an image descriptor for document classification and retrieval (Gabor, ~85% recall).
• Developed ML and classical CV algorithms, deploying models cross-platform via ONNX Runtime.

Software Engineer | IniVation AG | Zürich, CHFeb 2020 - Sep 2021

• Built event-based monocular depth estimation with <1mm accuracy and <10ms latency at 5kHz.

• Implemented data clustering and segmentation algorithms for neuromorphic sensor calibration.
• Reduced pipeline latency 30x to <10ms via profiling, OpenCL, SIMD and mathematical optimisation.

Embedded Systems Engineer | Siemens AG | Wallisellen, CHJul 2016 - Jan 2020

• Designed and evaluated slip/slide detection algorithms using Extended Kalman Filters, wavelet analysis, statistical tests, and SVMs.
• Developed SIL4 train protection software for Siemens Trainguard ETCS on-board unit, a platform deployed on 6,000+ trains across 50+ countries.

Embedded Software Engineer | Indel AG | Russikon, CHJul 2014 - Jun 2016

• Developed SIL3 functional safety firmware for industrial drive supervision, deployed on 100+ drives in production.
• Implemented multiprocessor synchronisation for drive position and velocity supervision.
• Designed PCBs for I/O testing of drive hardware.

Education

Theses and Academic Projects

Awards and Accolades

Appreciation Award issued by Electrosuisse AG for the Bachelor Thesis Piezoelectric Energy Harvesting for HSRvote

Natural Language Skills

Swiss German (native), German (bilingual), English (bilingual), French (elementary)