A

Architect Labs

AI-driven automated ASIC chip design platform

Microarchitect / RTL Design - Memory Subsystem

Silicon & ASIC DesignConfirmed live in the last 24 hours
Full-timePalo Alto, CA, USANo salary listedIn-person

About the Job

Requirements

  • A Bachelor's, Master's, or PhD in Electrical Engineering, Computer Engineering, or a closely related field.
  • At least 5 years of RTL design experience and at least one advanced-node tapeout involving memory subsystems such as DDR/LPDDR/HBM controllers, cache hierarchies, or memory-intensive SoC subsystems.
  • Deep familiarity with JEDEC memory standards, including DDR5/LPDDR5X command/address protocols, timing parameters, training sequences, and/or HBM2E/HBM3 pseudo-channel architecture, stack addressing, and interleaving schemes.
  • Hands-on experience designing or owning memory controller blocks, including command schedulers, bank state machines, refresh engines, read/write turnaround optimization, and PHY interface timing such as DFI or proprietary interfaces.
  • Experience with multi-level cache design, scratchpad controllers, or unified memory architectures with partitioning and quality-of-service mechanisms.
  • Ability to produce clear, synthesizable, lint-clean SystemVerilog RTL, with parameterization for multiple standards, modular channel or pseudo-channel instantiation, and configurable capacity and bandwidth targets.
  • Hands-on experience with SRAM controllers and arbiters, bank conflict resolution, address hashing or interleaving, ECC encode/decode engines, and high-bandwidth movement between on-chip and off-chip memory.
  • A solid grasp of synthesis, timing constraints, clock domain crossings, reset strategies, AMBA protocols (AXI, ACE, CHI), and power management for memory subsystems.
  • Strong Python skills for design automation, performance modeling, regression infrastructure, and tooling.
  • Experience taking a memory controller or cache subsystem from RTL through synthesis and working with physical-design teams on timing, area, and power closure, particularly for high-frequency controller logic and wide data buses.

Responsibilities

  • Own memory subsystem RTL end-to-end, from DDR/HBM controller design through code generation, lint, clock-domain-crossing checks, synthesis, and timing closure using the AI-driven design flow.
  • Design and implement memory controllers, including DDR5/LPDDR5X PHY-side controller logic, HBM3/HBM3E pseudo-channel controllers, command scheduling, refresh management, and ECC/RAS engines.
  • Architect the memory hierarchy, including multi-level cache controllers, scratchpad memory managers, coherency protocol engines where applicable, prefetch engines, and bandwidth partitioning and quality-of-service mechanisms for ML accelerator datapaths.
  • Design memory-side accelerators, including near-memory compute logic, scatter-gather DMA engines, address translation/remapping units, compression/decompression engines co-located with memory interfaces, and intelligent prefetchers tuned for ML access patterns.
  • Work with the principal architect to refine microarchitectural specifications, resolve bandwidth, latency, area, and power trade-offs, and feed area, timing, and power realities into the architecture and internal AI systems.
  • Define and maintain interface specifications for DDR PHY (DFI), HBM PHY, on-chip SRAM, AXI/ACE/CHI memory-facing fabric ports, and custom near-memory accelerator datapaths.
  • Build and maintain RTL infrastructure for the in-house AI-driven flow, including design automation scripts, regression flows, lint and clock-domain-crossing waivers, and memory subsystem integration collateral.
  • Support verification bring-up with memory timing models, protocol-compliant bus functional models, SystemVerilog Assertions for JEDEC compliance, coverage plans targeting worst-case scheduling scenarios, and architectural documentation for verification closure.
  • Support and guide software and machine-learning experts in revising and improving the in-house AI flow based on memory subsystem expertise, particularly workload-driven memory access pattern optimization.
  • Support FPGA prototyping on Xilinx for early functional validation of memory controllers, including bring-up with DDR Memory Interface Generator IPs and HBM validation platforms.

Desired Qualifications

  • 10 or more years of RTL design experience.
  • Experience with HBM integration, including interposer-level considerations, PHY calibration, and thermal-management impacts on refresh.
  • Familiarity with CXL memory pooling, Type 3 device controllers, or disaggregated memory architectures.
  • Near-memory or processing-in-memory design experience.
  • Low-power design techniques, including DVFS-aware memory scheduling, partial-array self-refresh, clock gating of idle channels, and power gating of unused banks.
  • FPGA prototyping experience with Xilinx Vivado/Vitis and DDR Memory Interface Generator or HBM subsystem IP integration.
  • SystemVerilog Assertions for JEDEC protocol compliance, including command sequencing, timing parameter checking, and training state machines.
  • Prior IP building and delivery experience for DDR/LPDDR controllers, HBM controllers, or cache subsystem IPs.
  • Experience building or using cycle-accurate memory system simulators such as DRAMSim or Ramulator to validate microarchitectural decisions.
  • Publications or patents in memory systems, memory scheduling algorithms, or memory-centric compute architectures for ML workloads.

About the Company

Architect Labs

AI-driven automated ASIC chip design platform

Architect Labs develops AI-powered tools to automate and speed up ASIC design for the semiconductor industry. It combines artificial intelligence with an integrated workflow for hardware exploration, SystemVerilog construction, verification, and physical design to produce end-to-end, verified chip designs. The platform aims to discover new architectures and generate ready-to-tute designs that are validated for modern machine learning workloads, enabling hardware/material co-optimization between AI models and specialized silicon. Compared to traditional EDA and ASIC design services, Architect Labs focuses on AI-guided architecture search, automated verification, and full-stack co-design to reduce time, cost, and labor in chip development. The goal is to shorten multi-year design timelines, lower upfront costs, and empower more specialized, high-performance silicon for data centers, automotive, robotics, and edge applications.

Company size
11-50
Company stage
Seed
Total funding
$24M
Headquarters
Palo Alto, CA
Founded
2025

Recently Posted Jobs

Explore more opportunities at Architect Labs

See more jobs