About Phinity
Phinity is building prompt-to-silicon: the agent infrastructure that enables AI agents to design, verify, optimize, and ultimately tape out chips. By 2028, any company will be able to specify a set of requirements for a chip and Phinity will deliver them a GDS file in weeks, not years.
Today, chip development requires large teams, fragmented tools, and multi-year development cycles. We are building a closed-loop system in which models can reason about hardware, use engineering tools, learn from verification feedback, and perform increasingly consequential parts of the chip-development process.
We work with leading frontier AI labs and are building a small team with deep experience across AI research, silicon architecture, design, and verification. Our long-term goal is autonomous tapeout, making custom silicon accessible enough to accelerate technological progress at breakneck speed.
The Role
The HW/SW Codesign Engineer role is for an engineer who works at the boundary between silicon and software and wants to define how a new chip actually gets used, heavily using AI to speed up development. Some example responsibilities include:
Design the hardware/software interface for edge compute chips: the programming model, command processors, memory movement, and runtime-visible scheduling that let software drive the hardware
Build and own the firmware, drivers, and runtime for chip bring-up and expose its capabilities to software
Make co-design tradeoffs across the boundary, deciding what belongs in hardware, firmware, or software, and own the technical bar for how they fit together
Validate the subsystem across emulation, FPGA, and bring-up, making sure software-visible behavior matches intent
Requirements
6+ years building across the HW/SW boundary: firmware, drivers, runtimes, or embedded systems, with a working understanding of the underlying silicon
Knowledge of how accelerators and SoCs are programmed: command and queue models, DMA and memory movement, scheduling, and instruction or ISA interfaces
Experience using AI to develop hardware or software
Strong communication skills and willingness to try new methods in ambiguous and fast-moving environments
Why you should work with us
→ Build a real chip with unlimited AI usage
→ Access non-publicly released foundation models and learn what the frontier of intelligence is before everyone else
→ Own the hardware/software interface for a novel chip end-to-end
We offer competitive salary and equity. We work in-person from SF and also have benefits including:
Health insurance
Retirement plans
Unlimited PTO
Paid parental leave
Lunch and dinner provided in-office
Skills Required
- 6+ years building across the HW/SW boundary (firmware, drivers, runtimes, or embedded systems) with working understanding of silicon
- Knowledge of accelerator and SoC programming: command/queue models, DMA, memory movement, scheduling, and ISA interfaces
- Experience using AI to develop hardware or software
- Strong communication skills and ability to work in ambiguous, fast-moving environments
What We Do
Phinity Labs is building foundational training infrastructure for autonomous chip design. Its AI agents are intended to design, verify, and tape out custom chips end-to-end, while the company develops environments and data to teach large language models hardware workflows. The mission is to overcome the scarcity of open chip-design training data and enable more efficient, domain-specialized accelerators and broader hardware automation.
