Member of Technical Staff - Research Scientist

Architect Labs•Palo Alto, CA

20h

About The Position

Architect Labs is a frontier AI lab focused on custom silicon. It partners with frontier labs, cloud and neocloud providers, physical AI companies, and advanced fabs to develop custom chips co-designed for next-generation AI workloads. The company aims to shorten end-to-end software-to-silicon timelines and maximize intelligence per watt and per dollar. As a Member of the Technical Staff, you will work at the intersection of chip design and frontier AI, applying hardware engineering expertise to agentic systems that can reason about, generate, and verify silicon. You will research self-improving harnesses that guide swarms of agents working autonomously over long periods, collaborate with ML research, post-training, and infrastructure teams, and build production software, evaluations, and environments grounded in real hardware engineering tasks. The role involves translating ambiguous chip-design challenges into concrete agent capabilities and iterating quickly in a 0-to-1 startup environment.

Requirements

PhD in Computer Science or a related field, or at least 3 years of research experience at a frontier AI lab.
Experience designing and improving agent harnesses, including orchestration, tool use, memory, feedback loops, and recovery mechanisms for complex, long-horizon autonomous tasks.
Ability to stay current with frontier AI research, assess new approaches critically, and turn promising findings into measurable improvements.
Exceptional systems engineering skills, including architecting and shipping reliable, large-scale infrastructure and maintaining research code to production standards.
Ability to own ambiguous problems end to end, prototype quickly, and turn promising ideas into production systems.
Pragmatic, resourceful, and comfortable moving fast.
Interest in applying frontier AI to chip design and willingness to learn from hardware engineers.

Nice To Haves

Research published at leading conferences such as ICLR, ICML, or NeurIPS.
Meaningful open-source contributions to coding agents, agent harnesses, or ML tooling, or experience building a harness from scratch.
Experience designing and deploying production multi-agent workflows and agents that run autonomously for days or weeks.
Experience building evaluations and benchmarks for complex engineering tasks.
Familiarity with development reward signals, post-training, or reinforcement learning methods for language models.
Familiarity with training or serving transformer-based models and building high-performance, large-scale ML systems.

Responsibilities

Research self-improving harnesses that guide swarms of agents to work autonomously for days or weeks.
Stay current with state-of-the-art frontier AI research.
Partner with ML research, post-training, and infrastructure teams to turn hardware domain expertise into reward signals, benchmarks, and training signals.
Build scalable and robust software stacks that turn ideas into production systems.
Build evaluations and environments for chip-design tasks that ground model behavior in real hardware engineering work.
Prototype, dogfood, test, and iterate quickly in a 0-to-1 environment.
Translate ambiguous chip-design challenges into concrete agent capabilities that can ship.