Microarchitect / RTL Design - Network Datapath & MAC Engine Design

Architect is a frontier AI lab for chip design. We build AI models and tools for on-demand custom ASICs at scale. Our goal is to co-design custom ASICs alongside evolving ML workloads, and enable a new era of domain-specific chips that unlock capabilities impossible with current hardware paradigms. Born out of Stanford Research, our team blends AI with Silicon with a founding team from Anthropic, Google DeepMind, Meta SuperIntelligence, xAI, Apple and Intel.

As a Founding Member of the Technical Staff on the RTL Design team at Architect, you’ll own the AI-driven microarchitecture and RTL design of high-performance networking subsystems going into production silicon. You will define, drive, and revise the block-level micro-architecture specification for Ethernet MAC engines, packet processing pipelines, and network datapath logic — ensuring line-rate performance, protocol compliance, and seamless SoC integration.

• Own the network datapath RTL end-to-end: from IEEE 802.3-compliant MAC engine design through code generation, lint, CDC, synthesis, and timing closure using our AI-driven design flow.

• Design and implement Ethernet MAC engines: including 802.3 MAC/PCS layers, FCS generation/checking, flow control (802.3x pause, PFC), auto-negotiation state machines, and rate adaptation logic for multi-speed operation (10G/25G/50G/100G/400G+).

• Architect packet processing datapaths: including Tx/Rx FIFO management, packet buffering and scheduling, credit-based flow control, descriptor management, and DMA interfaces for host-to-network data movement.

• Work directly with the principal architect to refine microarchitectural specs, resolve implementation trade-offs (latency vs. throughput vs. area), and feed area/timing/power realities back into the architecture and internal AI systems.

• Define and maintain interface specifications: MII/GMII/XGMII/CGMII variants, AXI-Stream for packet data, AXI for CSR/DMA, and custom sideband signaling for scheduling and QoS.

• Build and maintain RTL infrastructure for our in-house AI-driven flow: design automation scripts, regression flows, lint/CDC waivers, and integration collateral for the networking subsystem.

• Close collaboration with DV: Support verification bring-up with protocol-aware reference models, SVA assertions for protocol compliance, coverage plans targeting corner-case packet scenarios, and architectural documentation for verification closure.

• Close collaboration with SW and ML: Support and guide our SW and ML experts to revise and improve our in-house AI flow based on your networking domain expertise.

• Support FPGA prototyping on Xilinx for early functional validation of the network datapath, including interoperability testing with commercial PHYs and switches.

• Degree: Bachelor’s, Master’s, or PhD in Electrical Engineering, Computer Engineering, or a closely related field.

• Experience: 5+ years (10+ preferred) in RTL design with at least one advanced-node tapeout experience involving networking or packet-processing silicon (NICs, switches, SmartNICs, or networking SoC subsystems).

• 802.3 Expertise: Deep familiarity with IEEE 802.3 Ethernet standards — MAC sublayer operation, PCS encoding (64b/66b, Reed-Solomon FEC), reconciliation sublayer, and multi-lane distribution/alignment.

• MAC Engine Design: Hands-on experience designing or owning Ethernet MAC blocks including Tx/Rx pipelines, inter-packet gap management, preamble/SFD processing, FCS computation, VLAN tagging, and jumbo frame support.

• SystemVerilog: Clear, synthesizable, lint-clean RTL with strong design habits — parameterization for multi-rate support, modularity for protocol layering, and configurability for different deployment targets.

• Packet Datapath Depth: Hands-on experience with packet buffering architectures, credit-based and descriptor-based DMA engines, store-and-forward vs. cut-through trade-offs, and scheduling/arbitration for network QoS.

• SoC Methodology: Solid grasp of synthesis, timing constraints, clock domain crossings (especially between MAC clock domains and core clocks), reset strategies, AMBA protocols (AXI, AXI-Stream), and power management.

• Python: Strong skills for design automation, regression infrastructure, protocol compliance test generation, and tooling.

• PPA Ownership: Experience taking a networking block from RTL through synthesis and working with PD teams on timing/area/power closure — particularly for high-frequency datapath logic.

• Experience with high-speed SerDes integration and PHY-side interfacing (CAUI, 56G/112G PAM4 SerDes).

• Familiarity with RDMA, RoCEv2, or TCP offload engine architectures.

• Experience with traffic management: scheduling algorithms (WRR, DWRR, strict priority), shaping, and policing.

• Low-power design techniques: clock gating for idle ports, power gating for unused lanes, multi-voltage domains.

• FPGA prototyping experience (Xilinx Vivado/Vitis), especially with Ethernet hard IPs or soft MAC cores.

• SVA assertions for protocol compliance checking (e.g., inter-frame gap, pause quanta timing, alignment marker lock).

• Prior IP building and delivery experience for Ethernet MAC/PCS or NIC subsystem blocks.

• Domain-specific research contributions: publications or patents in network-on-chip, datacenter networking, or high-performance packet processing hardware.

You’ll join a founding team building the future of chip design at the intersection of AI and silicon. Your networking expertise will directly shape production ASICs and influence how AI transforms hardware development — from spec to tapeout.