Selected Work & Open-Source Architecture
Osiris (simplifaisoul/osiris)
Core Architect on Osiris — real-time 3D geospatial and multi-sensor intelligence dashboard built for high-frequency streaming telemetry.
pureQTern & project-zero
Memory-safe Rust and zero-dependency C inference runtime for 1.58-bit ternary models, eliminating heap jitter on constrained edge compute.
8b-is / MLX-QUANT
Low-bit ternary quantization and custom GPU kernels designed to fit high-parameter perception and planning models onto embedded automotive GPUs.
lissajoverse & enthea
Sub-millisecond phase/frequency signal processing and deterministic 60fps shader pipelines.
Kernel & Embedded Security Tooling
Production Linux kernel internals (kptr_restrict research, eBPF telemetry) and low-level C/Android instrumentation from Inpedio & Cygnet.
Osaurus & Headroom (PR #1418)
Core Architect on Osaurus (offline autonomous agent execution with cryptographic identity) and contributor to Headroom kompress routing.
Direct Portal, LinkedIn Contacts & 1-Click Outreach
Shinpei Kato (加藤真平)
Founder of TIER IV and Autoware, leading open-source autonomous driving OS architecture.
Kazuya Takeda (竹田一哉)
Co-founder leading human-centric autonomous mobility and sensor signal processing.
TIER IV Global Engineering Talent Acquisition
Global English-speaking engineering recruitment team in Shinagawa.
Hi! I built an interactive real-time LiDAR perception & Rust systems portfolio specifically for TIER IV's Autoware engineering team in Shinagawa: https://tier4.vaked.dev — combining zero-allocation C/Rust engines, edge GPU quantization & OSINT 3D architecture (Osiris). Would love to connect!
Dear TIER IV Autoware Engineering & Talent Team, I built a live interactive LiDAR perception simulation and bilingual engineering application package tailored for TIER IV in Shinagawa, Tokyo: • Live Interactive CV (Real-Time LiDAR Point Cloud & Trajectory Planner): https://tier4.vaked.dev • Formal 履歴書 + 職務経歴書 (PDF): https://tier4.vaked.dev/resume.pdf • Edge & Systems Portfolio (PDF): https://tier4.vaked.dev/portfolio.pdf • Full Application Dossier: https://tier4.vaked.dev/application I bring 9+ years of production systems engineering across C, Rust, Go, Linux kernel internals (Inpedio kptr_restrict & eBPF), real-time 3D spatial visualization (Core Architect on Osiris — simplifaisoul/osiris), and 12.80× compressed edge GPU inference (MLX-QUANT / pureQTern). I am ready to contribute immediately to Autoware's safety-critical Rust and edge perception stack in Tokyo. Warm regards, Péter József Lódri https://www.linkedin.com/in/peter-lodri/
Technical Stack & Capabilities
Rust, C, Swift 6, Go — SIMD, custom Metal/WebGL2 GPU kernels, BitNet b1.58 {-1, 0, +1} ternary quantization (12.80× memory reduction), zero-dependency binaries.
Unity, Three.js, WebGL2, GLSL/HLSL shaders, oscilloscope vector synthesis, LiDAR/geospatial 3D telemetry (Osiris), procedural Web Audio.
Core Architect on Osaurus (osaurus-ai/osaurus) & Osiris (simplifaisoul/osiris), contributor to Headroom (PR #1418 kompress router), creator of AgentField verifiable DAGs.
9+ years shipping high-concurrency Go/Python/Rust/TS systems (ASAPP, DailyWire 110k concurrent users, institutional fintech ETL). Fluent English, native Hungarian, conversational multi-lingual + daily Japanese in Tokyo.
Why TIER IV (ティアフォー · Autoware)
TIER IV's leadership of The Autoware Foundation proves that open-source, verifiable, safety-critical engineering is the only scalable path to autonomous mobility. In autonomous vehicles, heap jitter, memory bandwidth stalls, and opaque black-box pipelines cost lives. My career bridges low-level Linux kernel/C internals, memory-safe Rust, real-time 3D telemetry visualization (Osiris), and 12.80× compressed edge GPU inference (pureQTern / MLX-QUANT) — the exact toolkit needed to make Autoware faster, leaner, and provably deterministic.
TIER IVが主導するオープンソース自動運転OS「Autoware」は、安全性と透明性が求められる自動運転社会のグローバルスタンダードです。組み込みLinuxカーネルやC/Rustの低レイヤー実装、リアルタイム3D空間テレメトリ基盤（Osiris）、そして車載エッジGPUのメモリ帯域制約を打破する1.58ビット量子化技術（pureQTern / MLX-QUANT）を融合させ、品川のTIER IVチームでAutowareのRust化とリアルタイム知覚パイプラインの進化に貢献したいと強く志望しています。