Eridu AI Logo

Eridu AI

Software Engineer – High-Speed SerDes

Posted 14 Days Ago
Be an Early Applicant
In-Office
Saratoga, CA, USA
195K-265K Annually
Expert/Leader
In-Office
Saratoga, CA, USA
195K-265K Annually
Expert/Leader
Leads embedded software and firmware development for high-speed SerDes PHYs supporting Ethernet, PCIe, and UCIe interfaces. Responsibilities include PHY initialization, link training, equalization, calibration, diagnostics, characterization, production testing, health monitoring, and failure analysis. The role spans pre- and post-silicon bring-up, laboratory validation, system-level signal integrity, and optimization across operating conditions. It also defines reusable software architecture, creates technical documentation, mentors engineers, and serves as the technical authority for SerDes software and system bring-up.
The summary above was generated by AI
About Eridu

Eridu is a Silicon Valley-based hardware startup pioneering infrastructure solutions that accelerate AI data centers to deliver Faster AI. Today’s AI performance is frequently limited by communication bottlenecks. Eridu introduces multiple industry-first innovations across silicon, packaging, software, and systems to deliver an order of magnitude improvement in performance and unlock greater GPU utilization to speed training job completion times and tokens-per-second for more profitable inference. We do this while simultaneously reducing capital and power costs and improving reliability.

 

The company’s solutions and value proposition have been widely validated by leading hyperscalers.

 

Eridu has raised over $200M to date including its most recent, oversubscribed Series A round. The company is led by a veteran team of Silicon Valley executives who have delivered multiple billion dollar product lines and led multiple companies to billion dollar exits, including serial entrepreneur Drew Perkins, co-founder of Infinera (NASDAQ: INFN), Lightera (acq. by Ciena), Gainspeed (acq. by Nokia) and Mojo Vision (the world’s leading micro-LED company). The company is in execution mode and has a world-class engineering team with decades of experience in state-of-the-art silicon, packaging, optics, software, and systems. Eridu is working with best-in-class supply chain partners including silicon, packaging and systems.

 

Visit our website eridu.ai to learn more.


Position Overview

We are seeking a highly experienced SerDes software engineer to lead the development, bring-up, validation, and optimization of next-generation high-speed interfaces. The successful candidate will work with 224 Gb/s-per-lane Ethernet SerDes, PCI Express, and UCIe chiplet interconnects in advanced networking and compute silicon.

This is a senior technical role requiring strong knowledge of SerDes architecture, PHY initialization, link training, equalization, calibration, diagnostics, and system-level signal-integrity behavior. The engineer will work closely with analog and mixed-signal designers, ASIC architects, firmware teams, validation engineers, board designers, and system software developers.

Key Responsibilities
  • Develop embedded software and firmware for configuring, initializing, calibrating, and monitoring high-speed SerDes PHYs.
  • Lead bring-up and optimization of 224G PAM4 SerDes used in Ethernet switch, networking, and accelerator products.
  • Develop software support for PCIe SerDes, including link initialization, training, equalization, lane management, error recovery, and diagnostics.
  • Develop software support for UCIe die-to-die interfaces, including PHY initialization, link training, lane repair, health monitoring, and failure analysis.
  • Implement SerDes adaptation and calibration workflows, including:
    • Transmitter amplitude, pre-cursor, and post-cursor tuning
    • Receiver CTLE, VGA, DFE, and CDR configuration
    • Adaptive equalization
    • Eye optimization and margining
    • Lane-to-lane skew management
    • Polarity inversion and lane remapping
    • Background calibration and environmental compensation
  • Develop APIs, drivers, firmware interfaces, command-line tools, and diagnostic utilities for SerDes control.
  • Support Ethernet auto-negotiation and link training, including Clause 73 and Clause 72/136–related workflows where applicable.
  • Support PCIe LTSSM interaction, preset and coefficient negotiation, lane margining, retraining, and recovery.
  • Support UCIe link state management, PHY repair mechanisms, sideband interactions, and link reliability features.
  • Create automated characterization and production-test software for:
    • PRBS generation and checking
    • BER measurement
    • Eye scans and bathtub curves
    • Jitter and signal-margin analysis
    • Loopback testing
    • Lane failure isolation
    • Channel and package characterization
  • Analyze difficult link failures across silicon, package, board, connector, cable, and optical-module channels.
  • Correlate software-visible SerDes telemetry with oscilloscope, BERT, VNA, and simulation results.
  • Develop algorithms and policies for runtime SerDes health monitoring, telemetry, predictive failure detection, and fault recovery.
  • Optimize SerDes performance across process, voltage, temperature, channel loss, and manufacturing variation.
  • Define software architecture and reusable abstractions that support multiple SerDes IP vendors and PHY generations.
  • Create technical specifications, programming guides, bring-up procedures, debug playbooks, and validation plans.
  • Mentor engineers and serve as the organization’s technical authority for SerDes software and system bring-up.
Required Qualifications
  • Bachelor’s or Master’s degree in Electrical Engineering, Computer Engineering, Computer Science, or a related field.
  • Typically 10 or more years of experience in SerDes firmware, PHY software, silicon bring-up, or high-speed interface validation.
  • Deep understanding of high-speed serial-link fundamentals, including:
    • PAM4 and NRZ signaling
    • Channel loss, insertion loss, and return loss
    • Jitter, noise, crosstalk, and reflections
    • Clock and data recovery
    • Transmitter feed-forward equalization
    • CTLE and decision-feedback equalization
    • BER, eye diagrams, and link margin
    • Forward error correction
  • Hands-on experience bringing up and debugging at least one advanced SerDes generation, preferably 112G or 224G PAM4.
  • Strong experience with one or more of the following:
    • Ethernet SerDes and IEEE 802.3 link training
    • PCIe Gen5, Gen6, or later-generation PHYs
    • UCIe or another high-speed die-to-die interface
  • Strong embedded programming skills in C and C++.
  • Experience with Python or similar languages for automation, characterization, data analysis, and laboratory equipment control.
  • Experience programming SerDes registers through MDIO, I²C, SPI, PCIe configuration space, memory-mapped registers, or vendor-specific interfaces.
  • Familiarity with embedded processors, microcontrollers, real-time firmware, interrupts, DMA, and hardware/software interfaces.
  • Experience debugging pre-silicon models, emulation platforms, FPGA prototypes, and post-silicon devices.
  • Ability to interpret schematics, package and board topology, channel models, IBIS-AMI results, and signal-integrity reports.
  • Experience using laboratory equipment such as high-bandwidth oscilloscopes, BERTs, VNAs, protocol analyzers, and traffic generators.
  • Strong root-cause analysis skills and the ability to debug complex issues spanning analog, digital, firmware, package, board, and system domains.
Preferred Qualifications
  • Experience with 224G-per-lane Ethernet PHYs and 800G, 1.6T, or higher-speed Ethernet systems.
  • Experience with PCIe 6.x or PCIe 7.0, including PAM4 signaling, FLIT mode, FEC, and lane margining.
  • Experience with UCIe PHY and adapter-layer bring-up in multi-die or chiplet-based systems.
  • Familiarity with long-reach, medium-reach, short-reach, very-short-reach, and die-to-die SerDes channels.
  • Experience integrating third-party SerDes IP into an ASIC or SoC.
  • Experience developing vendor-independent SerDes abstraction layers.
  • Knowledge of CMIS, pluggable optics, retimers, active electrical cables, and linear-drive optics.
  • Familiarity with IEEE 802.3, PCI-SIG, and UCIe Consortium specifications.
  • Experience with SerDes power optimization, fast link recovery, degradation detection, and field diagnostics.
  • Experience developing production screening, yield-analysis, and manufacturing-test solutions.
  • Experience working with advanced process nodes, 2.5D packaging, interposers, and chiplet-based architectures.
  • Track record of technical leadership, cross-functional execution, and mentoring.
Success in This Role

The successful candidate will:

  • Bring up new SerDes interfaces quickly and methodically.
  • Establish stable links across challenging channels and operating conditions.
  • Convert low-level SerDes capabilities into robust, maintainable software.
  • Provide actionable diagnostics rather than opaque link-up or link-down status.
  • Reduce silicon bring-up and customer-debug time through automation and strong observability.
  • Build a reusable SerDes software architecture spanning Ethernet, PCIe, UCIe, and future high-speed interfaces.
Role Scope

This position focuses on SerDes control software, embedded firmware, link management, diagnostics, characterization, and system-level optimization. It will work closely with—but does not replace—the analog/mixed-signal SerDescircuit-design function.


Why Join Us? 

At Eridu, you’ll have the opportunity to shape the future of AI infrastructure, working with a world-class team on groundbreaking technology that pushes the boundaries of AI performance. Your contributions will directly impact the next generation of AI infrastructure solutions, transforming the performance of AI data centers. 

 

The starting base salary for the selected candidate will be established based on their relevant skills, experience, qualifications, work location, market trends, and the compensation of employees in comparable roles. 


Notice to Recruiting Agencies

Eridu does not accept unsolicited resumes or candidate profiles from staffing agencies or third-party recruiters. Any candidate submitted to Eridu without prior written authorization from our recruiting team will be considered unsolicited and will become the property of Eridu. Eridu reserves the right to pursue and hire such candidates without any obligation to pay fees. Recruiting agencies are expressly instructed not to contact hiring managers, employees, or executives regarding open positions.

HQ

Eridu AI Saratoga, California, USA Office

12980 Saratoga Ave, Saratoga, California, United States, 95070

Similar Jobs

21 Seconds Ago
Hybrid
Santa Clara, CA, USA
191K-334K Annually
Senior level
191K-334K Annually
Senior level
Artificial Intelligence • Cloud • HR Tech • Information Technology • Productivity • Software • Automation
Designs and operates cloud-native reliability, release, testing, and developer productivity platforms. Builds Kubernetes infrastructure, CI/CD and GitOps workflows, automated validation, observability, progressive delivery, resilience testing, and self-service engineering tools. Resolves complex platform and networking issues, improves release confidence and operational efficiency, partners across engineering teams, participates in architecture decisions, and mentors engineers. Requires extensive SRE, DevOps, platform, software, or infrastructure engineering experience and strong programming skills.
Top Skills: AnsibleArgo CdArgo WorkflowsAws EksAzure AksCi/CdCypressDockerFluxGateway ApiGitlab Ci/CdGitopsGoGoogle Cloud GkeHelmIngressIstioJavaJunitKubernetesKustomizeLinkerdOpentelemetryPlaywrightPrometheusPytestPythonRest AssuredRubySeleniumTerraformTestng
23 Seconds Ago
Hybrid
Mountain View, CA, USA
143K-243K Annually
Senior level
143K-243K Annually
Senior level
Artificial Intelligence • Cloud • HR Tech • Information Technology • Productivity • Software • Automation
Designs and builds core infrastructure for Moveworks, including distributed data stores, authentication, event streaming, configuration, rate limiting, feature flags, and observability systems. Partners with machine learning, search, product, data, and frontend teams to shape infrastructure strategy. Improves system reliability, scalability, latency, and performance through monitoring, bottleneck analysis, and distributed systems engineering.
Top Skills: AWSAzureC++Distributed SystemsDockerElasticsearchGCPGoIstioJavaKafkaMicroservicesOpensearchPython
24 Seconds Ago
Hybrid
Santa Clara, CA, USA
166K-290K Annually
Senior level
166K-290K Annually
Senior level
Artificial Intelligence • Cloud • HR Tech • Information Technology • Productivity • Software • Automation
Leads enterprise cloud FinOps across AWS, Azure, GCP, private infrastructure, and AI services. Develops governance, cost allocation, rate optimization, commitment strategies, savings pipelines, unit-economic models, forecasting, and executive reporting. Partners with Engineering, Finance, Product, Procurement, and cloud providers to improve infrastructure efficiency and gross margin. Helps shape AI-powered FinOps tooling, including anomaly detection, forecasting, optimization recommendations, and natural-language cost analysis. Mentors practitioners and advises senior stakeholders.
Top Skills: Artificial IntelligenceAWSGoogle Cloud PlatformAzureServicenowSQL

What you need to know about the San Francisco Tech Scene

San Francisco and the surrounding Bay Area attracts more startup funding than any other region in the world. Home to Stanford University and UC Berkeley, leading VC firms and several of the world’s most valuable companies, the Bay Area is the place to go for anyone looking to make it big in the tech industry. That said, San Francisco has a lot to offer beyond technology thanks to a thriving art and music scene, excellent food and a short drive to several of the country’s most beautiful recreational areas.

Key Facts About San Francisco Tech

  • Number of Tech Workers: 365,500; 13.9% of overall workforce (2024 CompTIA survey)
  • Major Tech Employers: Google, Apple, Salesforce, Meta
  • Key Industries: Artificial intelligence, cloud computing, fintech, consumer technology, software
  • Funding Landscape: $50.5 billion in venture capital funding in 2024 (Pitchbook)
  • Notable Investors: Sequoia Capital, Andreessen Horowitz, Bessemer Venture Partners, Greylock Partners, Khosla Ventures, Kleiner Perkins
  • Research Centers and Universities: Stanford University; University of California, Berkeley; University of San Francisco; Santa Clara University; Ames Research Center; Center for AI Safety; California Institute for Regenerative Medicine

Sign up now Access later

Create Free Account

Please log in or sign up to report this job.

Create Free Account