Phinitylabs
Architect
1w ago
USASeniorchip designmicroarchitectureverificationtiming analysispower optimizationphysical design
Architect and lead the development of edge compute chips using AI to optimize design and performance.
Responsibilities
- The Chip Architect role is for an experienced silicon engineer who wants to architect and lead a a chip, heavily using AI to speed up development. Some example responsibilities include:
- Architect and drive the development of edge compute chips, owning the microarchitecture and the PPA, timing, area, power, and physical-feasibility tradeoffs that define a good design
- Make the chip- and subsystem-level architecture decisions that have to survive verification, timing, area, power, and physical constraints, and own the technical bar across the design
- Establish rigorous verification and signoff practice across simulation, formal, and physical design
- Articulate why a design decision is good, and help us encode that judgment into the tooling that lets a small team design hardware at unusual speed
- Work hands-on with unreleased frontier models as part of how the chip gets built
- The role will involve you working directly with the founders and collaborating frequently with frontier lab researchers.
Requirements
- 8+ years architecting and shipping complex chips or subsystems, with designs taken through tapeout
- Knowledge across the design stack: RTL and microarchitecture, PPA optimization, verification, understanding of physical design and signoff
- Experience using AI to develop chips
- Strong communication skills and willingness to try new methods in ambiguous and fast-moving environments
Other
- Phinity is building prompt-to-silicon . We are creating the closed-loop system that lets AI agents design, verify, optimize, and tape out chips end-to-end. 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.
- We are working with leading frontier model labs and growing exponentially. Our team brings together deep AI and silicon expertise: training chip-design models at NVIDIA, doing formal reasoning post-training at scale, exiting a synthetic data company, leading the LPU team at Groq, and building chips at AMD, Intel, and MediaTek for over 10+ years. We are building an exceptional, highly efficient team to dismantle the current chip industry.
- → 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 architecture decisions for novel chip design 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