Software Engineer in Network Infrastructure
About the job
- At least 5 years of professional software engineering experience or equivalent practical background.
- Strong coding skills and an ownership mindset, with the ability to ship and operate reliable services.
- Proficiency in Go or readiness to switch to Go; Python is also welcome, and other languages may be useful for open-source debugging and fixes.
- Interest in infrastructure and/or networking; network expertise is not required.
- Ability to manage large-scale projects involving multiple stakeholders.
- Ability to break down complex tasks and guide personal work and that of more junior colleagues.
- Expertise in specific technologies and ability to write high-quality code that can serve as a reference.
- Ability to assess task priority and focus on high-impact work.
- Strong architectural thinking and ability to contribute to system design.
- Willingness to share knowledge and mentor others.
- Build and maintain services and tooling to automate the network lifecycle, including day-zero provisioning, day-N changes, drift detection, and operational verification.
- Make network changes safe and transparent through CI/CD workflows, diff and review tooling, staged rollouts and rollbacks, audit trails, and guardrails.
- Develop observability systems that scale across many sites, including telemetry pipelines, signal quality, and tooling that shortens incident investigations.
- Integrate source-of-truth data, expose APIs, and build reliable automation to close gaps between the network and the platform.
- Collaborate with network engineers and site reliability engineers to turn operational pain points into product-quality tooling.
- Assess task priority and focus on high-impact work.
- Contribute to system design and high-quality code.
- Participate in hiring and contribute to interviews.
- Share knowledge and mentor others.
- Background in networking, such as experience as a network engineer, CCNP, networking education, or data center networking exposure; alternatively, strong interest and proven ability to learn quickly.
- Experience building automation or infrastructure tooling, including CI/CD, infrastructure as code, testing or staging environments, or network-as-code workflows.
- Low-level networking or datapath experience, including eBPF/XDP, DPDK, kernel networking, or traffic-processing systems.
- Experience designing high-load services and observability platforms, including metrics, logs, traces, alerting, or regression detection.
- Open-source contributions or experience extending or debugging open-source components in production environments.
About the company
Nebius runs an AI cloud for developers, researchers, and enterprises that train models and put them into production. Its AI Cloud combines GPU computing, storage, networking, and managed services, while Token Factory runs models through an API so teams do not manage the underlying servers. Customers rent computing and storage at usage rates or reserve capacity, choosing resources that fit short experiments or sustained production workloads. Beyond the cloud, the group owns Avride, which develops autonomous vehicles and delivery robots, and TripleTen, which teaches technical skills for people changing careers. The publicly listed group is headquartered in Amsterdam, Netherlands.
Company Size
1,001-5,000
Company Stage
IPO
Headquarters
Amsterdam, Netherlands
Founded
1997
Get referred to Nebius
See people who can refer or advise you
Simplify's Take
What believers are saying
- Parent sales are expanding: August 2026 results showed quarterly group revenue rising 454% to $582.3 million.
- Enterprise distribution gains a channel: September 2026's Palantir partnership names Nebius its preferred sovereign AI infrastructure partner.
- Junior engineers get guided production experience: Early Talent participants work on real products with dedicated experienced buddies.
What critics are saying
- Existing shareholders faced dilution: August 2026 financing exchanged $800 million of older debt for approximately 15.8 million shares.
- Remote flexibility is limited: its careers FAQ usually permits only one or two home working days each week.
- Expansion still loses money: August 2026 results recorded a $175.9 million quarterly group operating loss.
What makes Nebius unique
- Hardware and software share one design: Nebius engineers its servers, racks, cooling, and cloud platform together.
- Computing heats nearby homes: its Finnish data center feeds recovered heat into the local district heating network.
- Its engineers started together: the founding team brought years of shared big-tech infrastructure work into Nebius.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Health Insurance
401(k) Company Match
Parental Leave
Remote Work Options
Disability Insurance
Life Insurance
Hybrid Work Options
Professional Development Budget
Growth & Insights and Company News
6 month growth
5%1 year growth
-1%2 year growth
-3%Nebius, a Netherlands-based AI infrastructure company, has seen its stock surge from $14.29 at its October 2024 debut to around $220 currently, after peaking at $286.69 in June 2026. The company, formerly known as Yandex, transformed into a "neocloud" provider after divesting its Russian assets following 2022 sanctions. Nebius operates AI-focused data centers across Finland, Iceland, and the US, with 14 more planned. Revenue jumped 351% to $530 million in 2025, whilst first-half 2026 revenue soared 529% year-over-year to $981 million. The company's contracted backlog exceeded $40 billion by mid-2026. Analysts project 2027 revenue of $12.3 billion and adjusted EBITDA of $6.6 billion. Trading at 8 times forward sales, the stock appears undervalued despite concerns about profitability and rising debt from its capital-intensive expansion strategy.
Jim Cramer expressed a preference for CoreWeave over Nebius on Mad Money, citing concerns about Nebius's recent price movement. The valuation gap between the two AI infrastructure companies is substantial. CoreWeave generated $2.575 billion in second-quarter revenue, up 112% year-over-year, compared with Nebius's $582.3 million, which grew 454%. CoreWeave reported approximately $104 billion in revenue backlog. Nebius trades at 46.30 times trailing sales and 47.03 times enterprise value to revenue. CoreWeave trades at 5.97 times sales and 12.38 times enterprise value to revenue, making Nebius's valuation multiples roughly four to seven times higher. Both companies face capital intensity challenges. CoreWeave carried $35.6 billion in total debt as of 30 June, whilst Nebius recorded $8.13 billion in property and equipment purchases during the first half. Hedge fund ownership increased for both companies in the second quarter.
Nebius shares have surged 181% this year to $235.88, driven by a $27 billion five-year deal with Meta and a $2 billion equity investment from NVIDIA. The AI cloud company holds a $37.5 billion backlog and reported second-quarter revenue growth of 454%. However, significant risks remain. Three customers account for 59% of revenue, and the company faces capital expenditure guidance of $20-25 billion. Convertible notes carry a fair value of $20.8 billion, creating substantial dilution risk. Second-quarter capex of $5.66 billion far exceeded operating cash flow of $2.25 billion. Management expects over $9 billion in customer prepayments this year, with recent deals covering 50-60% of associated capex. Analysts see 20% upside to $284, but the stock trades 21% below its 52-week high. Execution remains critical as the company must convert deployed capacity into revenue before its next financing round.
Nebius has signed a 12-year agreement with AIB Data Centers for 50 MW of critical IT capacity at a site in the southeastern US. The deal includes optional renewals, though financial terms were not disclosed. AIB said customer prepayments, along with project-level debt and preferred equity, should fund a substantial portion of initial development costs. The capacity is backed by a 15-year electric service agreement for 65 MW of utility load and will be delivered across two data halls. Nebius provides AI cloud infrastructure for developers and enterprises. Andrey Korolenko, the company's chief product and infrastructure officer, said time-to-power is the binding constraint on AI infrastructure today.
Two AI infrastructure companies, Innodata and Nebius Group, offer contrasting investment profiles as demand for generative AI capabilities accelerates. Innodata provides data engineering and human expertise for AI model training, serving five of the "Magnificent Seven" tech giants. The company reported FY 2025 revenue of nearly $252 million, up 48% year-over-year, with net income of approximately $32.2 million. It operates debt-free with strong cash generation. Nebius Group operates cloud infrastructure for AI model development. The former Yandex unit pivoted to AI in 2025, achieving explosive growth with FY 2025 revenue reaching nearly $530 million, up roughly 350% year-over-year. Net income was approximately $102 million, though the company posted a free cash flow loss of nearly $3.7 billion due to capital spending. Innodata faces customer concentration risk, with one client accounting for nearly 58% of 2025 revenue. Nebius competes against hyperscalers like Amazon and Microsoft whilst managing capital-intensive infrastructure expansion. Despite a higher price-to-sales ratio of 42.6x versus Innodata's 7.3x, Nebius's lower forward P/E of 37.8x and projected $3.3 billion in 2026 sales present stronger growth prospects.