Top Tech Jobs & Startup Jobs

Lead and build Neurophos' in-house IT function across Sunnyvale and Austin. Own on-prem Linux compute for EDA/CAD, storage, backup, SSO/IdP (Okta), GitHub Enterprise, cloud (Azure) administration, network design, MSP governance, endpoint fleet management, and insourcing roadmap. Design hybrid connectivity, security/compliance evidence, and scale infrastructure from ~50 to 150+ employees while hiring and mentoring the IT team.

Perform die- and wafer-level optical characterization of metasurface photonic devices. Execute spectral and electro-optic measurements, automate instrument control and data acquisition with Python, analyze large datasets to assess yield and device performance, document campaigns, and improve test setups and measurement throughput.

Lead development of inference servers and racks, owning electrical, mechanical, and systems engineering from boards and interconnects through rack-level power and high-speed channel closure. Build and lead the Systems team, drive integration/bring-up, EVT/DVT/PVT sign-off, system reliability (RAS), and manage schedule and budget for full-scale product delivery.

Lead optical architecture, design, simulation, and optimization of free-space optical systems for an optical processing unit. Use Zemax/LightTools for tolerance, stray light, and sensitivity analysis; collaborate with optomechanical, photonics, electrical, and suppliers; support lab alignment, characterization, and system reviews; communicate performance, requirements, and risks to engineers and executives.

Lead software program execution to keep software delivery ahead of hardware by operating hybrid scrum-fall processes, running agile ceremonies, producing forward-looking forecasts with confidence levels, surfacing and driving technical and schedule risks to closure, and coordinating cross-functional dependencies across software, hardware, and cloud while advising chief engineers and executives on scope and sequencing.

Lead development of a compiler toolchain for a custom LLM accelerator: design IR/graph transformations and lowering passes, optimize compute and memory patterns, integrate with PyTorch/JAX/Triton, build test infrastructure, and ensure correctness and performance for hybrid optical-electronic hardware.