ABOUT ASTRUS

ABOUT ASTRUS

ABOUT ASTRUS

📍 Location: Toronto or Waterloo, Canada


At Astrus, we are using AI to automate microchip design, starting with the biggest bottleneck, analog layout. Our mission is to radically improve global computation and empower chip designers to create the world's most advanced microchips with AI. Astrus is backed by top-tier VC firms: Khosla Ventures, HOF Capital, and 1517 Fund.

📍 Location: Toronto or Waterloo, Canada


At Astrus, we are using AI to automate microchip design, starting with the biggest bottleneck, analog layout. Our mission is to radically improve global computation and empower chip designers to create the world's most advanced microchips with AI. Astrus is backed by top-tier VC firms: Khosla Ventures, HOF Capital, and 1517 Fund.

THE MISSION

THE MISSION

THE MISSION

Astrus is turning analog chip layout into a game that AI can learn to master.

We’re looking for a Senior / Principal Engineer to build the engine behind that game — the system that defines how layouts are represented, evaluated, and evolved at scale.

This is not about implementing predefined mechanics. You will help define the rules of the system itself: how the environment behaves, how it scales, and how it stays grounded in real-world semiconductor constraints.

You do not need prior chip design experience. What matters is your ability to build complex, high-performance systems. You’ll learn the domain as you go, working closely with experts — and your systems instincts are far more important than prior semiconductor knowledge.

You’ll work at the intersection of systems engineering, simulation, and AI, building an engine that powers both large-scale training and real-world inference.

If you’re excited by designing core systems, pushing performance limits, and turning messy, ambiguous problems into clean abstractions — this role is for you.

Astrus is turning analog chip layout into a game that AI can learn to master.

We’re looking for a Senior / Principal Engineer to build the engine behind that game — the system that defines how layouts are represented, evaluated, and evolved at scale.

This is not about implementing predefined mechanics. You will help define the rules of the system itself: how the environment behaves, how it scales, and how it stays grounded in real-world semiconductor constraints.

You do not need prior chip design experience. What matters is your ability to build complex, high-performance systems. You’ll learn the domain as you go, working closely with experts — and your systems instincts are far more important than prior semiconductor knowledge.

You’ll work at the intersection of systems engineering, simulation, and AI, building an engine that powers both large-scale training and real-world inference.

If you’re excited by designing core systems, pushing performance limits, and turning messy, ambiguous problems into clean abstractions — this role is for you.

WHAT YOU'LL OWN

WHAT YOU'LL OWN

WHAT YOU'LL OWN

  • The core engine. Design and evolve the Layout Game Engine that powers training, data generation, and inference across Astrus.

  • The interface between research and production. Work directly with Researchers to turn new modeling ideas into robust, scalable, and debuggable systems.

  • Performance at scale. Identify and eliminate bottlenecks across the system — improving throughput, latency, and cost efficiency for large-scale workloads.

  • System architecture and abstractions. Build clean, extensible designs that can support new PDKs, layout strategies, and increasingly sophisticated agent behaviors without collapsing under complexity.

  • Selective acceleration where it matters. Help evolve a Python-first system with targeted use of Rust, JAX, or other high-performance tools when needed.

  • High-leverage technical initiatives. Lead efforts that meaningfully move the system forward — whether that’s redesigning core subsystems, improving debuggability, or scaling evaluation pipelines.

  • Engineering foundations. Build the tooling that makes the engine reliable and usable: profiling, testing, observability, and clear interfaces between research and production systems.

  • The core engine. Design and evolve the Layout Game Engine that powers training, data generation, and inference across Astrus.

  • The interface between research and production. Work directly with Researchers to turn new modeling ideas into robust, scalable, and debuggable systems.

  • Performance at scale. Identify and eliminate bottlenecks across the system — improving throughput, latency, and cost efficiency for large-scale workloads.

  • System architecture and abstractions. Build clean, extensible designs that can support new PDKs, layout strategies, and increasingly sophisticated agent behaviors without collapsing under complexity.

  • Selective acceleration where it matters. Help evolve a Python-first system with targeted use of Rust, JAX, or other high-performance tools when needed.

  • High-leverage technical initiatives. Lead efforts that meaningfully move the system forward — whether that’s redesigning core subsystems, improving debuggability, or scaling evaluation pipelines.

  • Engineering foundations. Build the tooling that makes the engine reliable and usable: profiling, testing, observability, and clear interfaces between research and production systems.

WHAT MAKES THIS ROLE DIFFERENT

WHAT MAKES THIS ROLE DIFFERENT

WHAT MAKES THIS ROLE DIFFERENT

  • You’re building the “physics” of the system. This isn’t application code — it’s the engine that defines how the problem itself is represented and solved.

  • You’ll work on systems that must be both fast and correct. Every design decision balances performance, accuracy, and real-world constraints.

  • You operate where research meets production. You’ll turn evolving ideas into systems that actually run at scale.

  • You’ll tackle non-standard problems. Analog layout, reinforcement-style environments, and simulation-heavy workloads create challenges you won’t find in typical backend roles.

  • You’re building the “physics” of the system. This isn’t application code — it’s the engine that defines how the problem itself is represented and solved.

  • You’ll work on systems that must be both fast and correct. Every design decision balances performance, accuracy, and real-world constraints.

  • You operate where research meets production. You’ll turn evolving ideas into systems that actually run at scale.

  • You’ll tackle non-standard problems. Analog layout, reinforcement-style environments, and simulation-heavy workloads create challenges you won’t find in typical backend roles.

WHAT YOU BRING

WHAT YOU BRING

WHAT YOU BRING

Core Experience

  • You’ve built and maintained complex, performance-sensitive systems in production.

  • You’re fluent in at least one serious programming language (Python is ideal, but not required).

  • You have experience working closely with research or AI teams, where requirements evolve quickly.

  • You’re comfortable turning ambiguous problems into clean system designs and working implementations.

  • You’ve improved real systems through profiling, architecture changes, parallelism, caching, or low-level optimization.

  • You take ownership across the full lifecycle: prototype → production → scale → long-term evolution.

  • You’re willing to learn the fundamentals of analog layout and semiconductor constraints.

Senior vs Principal

  • Senior: You’ve independently owned and delivered major systems or subsystems.

  • Principal: You set technical direction, lead critical initiatives, and raise the bar across a team.

Bonus

  • Experience with JAX or numerical/simulation-heavy systems

  • Experience with Rust, C++, or other performance-oriented languages

  • Background in reinforcement learning, search, or simulation environments

  • Experience building engines, compilers, or stateful systems

  • Familiarity with semiconductor concepts (PDKs, layout, DRC/LVS, etc.)

  • Experience supporting both offline training and online inference from the same core system

  • Strong technical instincts and a bias toward proactively shaping direction

Core Experience

  • You’ve built and maintained complex, performance-sensitive systems in production.

  • You’re fluent in at least one serious programming language (Python is ideal, but not required).

  • You have experience working closely with research or AI teams, where requirements evolve quickly.

  • You’re comfortable turning ambiguous problems into clean system designs and working implementations.

  • You’ve improved real systems through profiling, architecture changes, parallelism, caching, or low-level optimization.

  • You take ownership across the full lifecycle: prototype → production → scale → long-term evolution.

  • You’re willing to learn the fundamentals of analog layout and semiconductor constraints.

Senior vs Principal

  • Senior: You’ve independently owned and delivered major systems or subsystems.

  • Principal: You set technical direction, lead critical initiatives, and raise the bar across a team.

Bonus

  • Experience with JAX or numerical/simulation-heavy systems

  • Experience with Rust, C++, or other performance-oriented languages

  • Background in reinforcement learning, search, or simulation environments

  • Experience building engines, compilers, or stateful systems

  • Familiarity with semiconductor concepts (PDKs, layout, DRC/LVS, etc.)

  • Experience supporting both offline training and online inference from the same core system

  • Strong technical instincts and a bias toward proactively shaping direction

WHY THIS ROLE IS EXCITING

WHY THIS ROLE IS EXCITING

WHY THIS ROLE IS EXCITING

You’ll be building a system that sits at the core of a new approach to chip design — where AI interacts with a structured environment that you help define.

This is a role for engineers who care about:

  • Deep systems design

  • Performance and correctness at scale

  • Clean abstractions over complex domains

  • Working on problems that don’t already have standard solutions

If you want to build something foundational — not just features on top — we should talk.

Email Talent@Astrus.ai

TRANSPARENCY

  • AI Disclosure: Astrus uses AI to assist with parts of the screening and assessment process. Final hiring decisions are made by humans.

You’ll be building a system that sits at the core of a new approach to chip design — where AI interacts with a structured environment that you help define.

This is a role for engineers who care about:

  • Deep systems design

  • Performance and correctness at scale

  • Clean abstractions over complex domains

  • Working on problems that don’t already have standard solutions

If you want to build something foundational — not just features on top — we should talk.

Email Talent@Astrus.ai

TRANSPARENCY

  • AI Disclosure: Astrus uses AI to assist with parts of the screening and assessment process. Final hiring decisions are made by humans.

Ready to radically improve global computation? 🚀📈🌎 🤖

Ready to radically improve global computation? 🚀📈🌎 🤖

Reach out to Talent@Astrus.ai for more details

Reach out to Talent@Astrus.ai for more details