C++ Software Engineer
About the job
- Strong programming skills in at least one language, ideally C++.
- Extensive knowledge of Linux kernel internals, containerization technologies, and virtualization.
- Deep understanding of workload and network isolation techniques in multi-tenant environments.
- Experience securing and hardening cloud infrastructure, particularly in environments with untrusted workloads.
- Strong background in workload and network isolation, network security, and cloud-native security practices.
- Experience with GPU programming and understanding of GPU-specific security concerns.
- Expand and extend the GPU cloud daemon.
- Design and deploy market-based resource management systems.
- Harden code and infrastructure to meet zero-trust standards.
- Benchmark, profile, and eliminate bottlenecks across hypervisor, container, and network layers.
About the company
Vast.ai runs a marketplace that connects GPU owners with users who need high-performance computing for AI and machine learning. Users browse available hardware, compare performance using the DLPerf scoring function, and rent compute through interruptible instances and spot auctions to save money. The platform aggregates offerings from individuals, data centers, and large providers, enabling a diverse, competitive marketplace while prioritizing security and regulatory compliance. This approach lets customers choose hardware that matches their needs and budget, rather than sticking to traditional cloud providers. Vast.ai’s goal is to make AI development more accessible and to improve the utilization of existing computing resources, contributing to cost efficiency and sustainability in tech.
Company Size
11-50
Company Stage
N/A
Total Funding
N/A
Headquarters
Los Angeles, California
Founded
2018
Get referred to Vast.ai
See people who can refer or advise you
Simplify's Take
What believers are saying
- Hundreds of B200s went live in August 2026, expanding premium supply quickly.
- May and August 2026 updates added templates, serverless endpoints, and Hugging Face bucket integrations.
- SOC 2 Type II and SOC 3 unlock regulated buyers that reject informal GPU marketplaces.
What critics are saying
- The marketplace depends on host uptime; one powered-off rig kills a renter’s workload.
- CopyFail mitigations in May 2026 show distributed security bugs can hit the whole fleet.
- RunPod, Lambda, and CoreWeave keep tightening pricing and enterprise features, compressing Vast.ai’s moat.
What makes Vast.ai unique
- Vast.ai combines peer-to-peer GPU supply with real-time bidding and per-second billing.
- Its May 2026 benchmarking CLI prices workloads on live H100s, A100s, 5090s, and 4090s.
- By August 2026, Vast.ai added SOC 2 Type II and a Trust Center.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Health Insurance
Dental Insurance
Vision Insurance
Life Insurance
401(k) Company Match
Company Equity
Company News
Clichmont AI announces public sale of $CLAI as European GPU Infrastructure expands. September 21, 2026 14:19 ET | Source: Clichmont ROAD TOWN, British Virgin Islands, Sept. 21, 2026 (GLOBE NEWSWIRE) - Clichmont AI has opened the public sale of $CLAI, the utility token at the center of its AI computing ecosystem, as the company expands its European GPU infrastructure across Norway and Spain. The 45-day public sale runs through October 30, 2026, with $CLAI available through Clichmont AI's platform at clichmont.ai. The public sale opened on September 16 at 5pm CET and will close on October 30 at 11:59pm CET. Eligible participants in supported jurisdictions can purchase $CLAI using USDT or USDC on BNB Smart Chain (BEP-20). The public-sale price is $0.135 per $CLAI, with a minimum purchase of $10 and no stated maximum per participant. The token price may change during the public sale. Building European GPU Infrastructure Clichmont is developing a European network of GPU infrastructure designed to serve artificial intelligence, machine learning, rendering, scientific computing and other GPU-intensive workloads. The company currently operates 12 high-performance GPUs in Billingstad, Norway, including eight NVIDIA H20 GPUs and four NVIDIA RTX PRO 6000 GPUs. In addition, 20 GPUs - 10 NVIDIA RTX 4090 and 10 NVIDIA RTX 5090 units - are located at the Port of Bodø and ready for the next stage of deployment. Clichmont is also deploying a further 40 NVIDIA RTX 5090 GPUs as part of its expanding infrastructure. Together, these deployments represent 72 GPUs across Clichmont's current operating and deployment pipeline. Clichmont operates primarily as a B2B infrastructure provider, offering dedicated GPU computing capacity and data-center solutions to businesses, institutions and professional customers. By operating its own infrastructure, the company maintains direct control over hardware, power, cooling and capacity planning. The model is complemented by Clichmont's partnership with Vast.ai, which provides marketplace access to distributed GPU computing capacity for a broader user base. Connecting Compute Infrastructure With $CLAI $CLAI is designed as a utility token within the Clichmont AI ecosystem, connecting the company's physical computing infrastructure with a digital ecosystem. Token-related mechanisms include staking opportunities, preferential access and discounts on Clichmont computing services, the βCLAI mechanism and participation in the project's buyback-and-burn model. Clichmont has established a minimum monthly allocation of $60,000 for $CLAI buybacks. Tokens acquired through the mechanism are divided between protocol-owned liquidity and permanent token burns. The company also plans to allocate 5% of revenue from each new Clichmont data center to the $CLAI token ecosystem through the buyback mechanism. This is intended to link the expansion of Clichmont's physical infrastructure with the development of the token ecosystem. Roadmap for the Next 12 Months Following the public sale, Clichmont has outlined several infrastructure and ecosystem milestones. On October 30, 2026, the company plans to begin preparations for fundraising for its second data center in Bodø, led by partner Cyril Greco. On December 1, 2026, $CLAI is scheduled for its first decentralized exchange (DEX) listing. On May 1, 2027, Clichmont plans to begin construction of its Alicante data center, further expanding its European AI computing infrastructure. The company expects to continue deploying GPU capacity while developing the broader $CLAI ecosystem alongside its physical infrastructure. "We are launching $CLAI as Clichmont moves from building the foundations of its infrastructure to deploying computing capacity and expanding its data-center network. We wanted the token to be connected to that real-world growth rather than exist as a standalone crypto project. As our compute activity and revenues increase, the $CLAI ecosystem is designed to develop alongside the company's infrastructure, including through the monthly buyback mechanism," said Alexis Cathalifaud, Founder and CEO of Clichmont. About Clichmont Clichmont is a B2B technology and infrastructure company focused on the development and operation of high-performance GPU computing infrastructure and data centers. The company provides compute capacity for artificial intelligence, machine learning, rendering, scientific computing and other GPU-intensive enterprise workloads. Clichmont is developing a European network of compute infrastructure while building the $CLAI ecosystem to connect its physical computing activity with a utility-driven digital economy. Media сontact Alexis Cathalifaud Founder & CEO, Clichmont [email protected] Official Channels Website: www.clichmont.com Clichmont AI: www.clichmont.ai $CLAI Platform: app.clichmont.ai X: @Clichmont Disclaimer:This sponsored content is provided by the content provider and does not necessarily reflect the views of this media platform or its publisher. The information is shared for general informational purposes only and should not be considered financial, investment, or trading advice. Cryptocurrency and mining-related activities carry risks, including the potential loss of capital, and readers are encouraged to conduct their own research and seek professional advice where appropriate. Speculate only with funds that you can afford to lose.The media platform and publisher assume no responsibility for any losses or claims arising from reliance on this content. GlobeNewswire does not endorse any content on this page. Legal Disclaimer: This article is provided on an "as-is" basis, without warranties or representations of any kind, express or implied. The media platform assumes no responsibility or liability for the accuracy, content, completeness, legality, or reliability of the information presented. Any complaints, claims, or copyright concerns related to this article should be directed to the content provider mentioned above.
GPU Mart review 2026: affordable GPU cloud servers for AI & ML. GPU Mart is a specialized GPU cloud hosting brand under Database Mart LLC, offering on-demand and reserved NVIDIA GPU servers for AI, machine learning, and rendering workloads. Here's its 2026 independent review. About GPU Mart. GPU Mart focuses exclusively on GPU cloud computing, with infrastructure in Dallas, TX and Kansas City, MO. It's part of the Database Mart LLC family, sharing the same SOC 2 Type II and ISO/IEC 27001 certified infrastructure as VPS Mart and Server Mart. GPU Mart pricing. On-Demand GPU plans. | GPU | VRAM | RAM | Storage | Hourly | Monthly | | NVIDIA RTX 3090 | 24 GB | 64 GB | 500 GB NVMe | $0.49/hr | $199/mo | | NVIDIA RTX 4090 | 24 GB | 64 GB | 500 GB NVMe | $0.89/hr | $349/mo | | NVIDIA A100 40GB | 40 GB | 128 GB | 1 TB NVMe | $2.49/hr | $999/mo | | NVIDIA H100 80GB | 80 GB | 256 GB | 2 TB NVMe | $3.99/hr | $1,499/mo | All plans include: CUDA pre-installed, PyTorch + TensorFlow images, Jupyter Notebook support, and NVMe-backed storage. Use cases. * AI/LLM training - H100 and A100 clusters for foundation model training * Inference workloads - Cost-efficient RTX 4090 deployments * 3D rendering - Blender, Cinema 4D, and Unreal Engine rendering farms * Scientific computing - CUDA-accelerated simulations Performance benchmarks. Its MLPerf benchmark on the H100 plan achieved 420 tokens/second on LLaMA 2 70B inference - competitive with top-tier GPU cloud providers at a significantly lower monthly cost. GPU Mart vs. Competitors. Compared to Lambda Labs, Vast.ai, and CoreWeave, GPU Mart offers predictable monthly pricing with no marketplace variability. For production AI workloads, this reliability advantage is significant. Verdict. GPU Mart is an excellent choice for developers and businesses needing reliable, affordable GPU cloud hosting. U.S.-based infrastructure, enterprise certifications, and competitive hourly/monthly pricing make it a strong alternative to mainstream GPU cloud providers. Rating: 4.6/5
Vast raises nearly $200M to become China's latest AI unicorn. Jun 01, 2026 - 07:04 A startup that didn't exist until 2023 just raised nearly $200 million and crossed the billion-dollar valuation threshold. Vast, a Beijing-based company building AI-powered 3D modeling tools, has become China's newest AI unicorn, adding another name to an increasingly crowded roster of generative AI companies jockeying for dominance in the world's second-largest economy. The company's core product, Tripo AI, translates text and image prompts into detailed 3D objects. That pitch has resonated with some serious names: clients include NetEase and Sony. The funding trail. Vast's path to unicorn status has been unusually fast, even by Chinese AI standards. The company closed a $50 million Series A round in March 2026, led by Alibaba Group and Hengxu Capital, with participation from Baidu Ventures and other investors. That single round pushed the company's valuation past the billion-dollar mark. Prior to that milestone, the startup had already pulled in tens of millions from a mix of government-linked and private investors, including the Beijing Artificial Intelligence Industry Investment Fund, Vitalbridge, and Fortune Capital. When you stack all the rounds together, the total funding approaches the $200 million figure. Who's behind it. Vast was founded by CEO Simon Song, who previously co-founded MiniMax, another Chinese AI unicorn. The company's user base tells its own story about product-market fit. Vast claims nearly 10 million individual users and roughly 90,000 studios and companies on the Tripo platform. The presence of enterprise clients like NetEase and Sony suggests the output quality is production-grade, or at least close enough that major studios are willing to integrate it into their workflows. The competitive landscape and what investors should watch. Vast positions itself as a competitor to efforts from Tencent and Google in text-to-3D generation. Tencent has the advantage of owning massive gaming and entertainment platforms that are natural distribution channels for 3D content. Disclosure: This article was edited by Editorial Team. For more information on how Ranzware inc. create and review content, see its Editorial Policy.
May 2026 product update. May 12, 2026 By Team Vast This month's updates bring new templates and guides, major upgrades to two-factor authentication, a benchmarking tool to help you pick the right GPU for your workload, and a complete creative AI studio that runs in a single container. NVIDIA Cloud GPU updates. Vast.ai Inc. has introduced several platform-level improvements this month. Two-factor authentication (2FA) now supports authenticator apps like Google Authenticator and 1Password instead of relying solely on SMS. Additional upgrades include CLI 2FA support, automatic backup codes for account recovery, multiple authentication methods per account, and 2FA for team roles and API key permission groups. Note: Action is required if you use legacy SMS 2FA. Simply log in normally, go to the Settings page, and click "Regenerate" on the backup codes section to migrate to the new 2FA in less than a minute. For developers, the Vast pip package has been upgraded, and the SDK and CLI repositories are now merged into a single, easy-to-grab repo. Vast.ai Inc. has also introduced an agent skill for coding assistants like Claude and Codex to drive its CLI more effectively. Highlighted feature: benchmark before you deploy. Choosing the right production GPU is now faster and easier, without any guesswork involved. Vast's benchmarking CLI will take your workload and test it on H100s, A100s, 5090s, 4090s, or any selection you define to compare performance and cost efficiency across GPU classes. The CLI rents one instance of each GPU type you specify, or a default set for a reasonable starting point if you don't specify any, and then runs the template's built-in benchmark workload, reports performance per dollar, and tears down each rental automatically when finished. Multiple GPU types can be tested in parallel, making it much faster to evaluate your options before committing to longer rentals or production traffic. Because each run uses real instances and consumes account credit, you get measured performance and cost per hour on actual hardware, not estimates. This benchmarking tool is built for the moment you're staring at a pricing page, wondering whether an H100 is actually worth 4x the cost of a 5090. Sometimes it is, and sometimes it isn't. Now you can find out before you deploy. New templates and guides. Vast.ai Inc. has rolled out another batch of templates and guides to support creative workflows, fine-tuning, experimentation, and multimodal reasoning. The standout addition this month is the All-in-One App Studio template. It packs eight creative AI applications and a GPU-accelerated remote desktop with KDE Plasma and Blender into a single container, giving you a complete AI production environment without having to manage separate instances. New templates. * All-in-One Creative AI Studio - One GPU instance: image generation, video, music, voice, transcription, LoRA training, LLM fine-tuning, and KDE desktop with Blender, each activated only when you need it. * Unsloth Studio - No-code UI for running and fine-tuning 500+ open-source models with 70% less VRAM. * Kimi K2.6 - 1T-parameter MoE, 32B active, multimodal agentic model with 256K context. * Qwen3.5 27B and Qwen 3.5B-A3B - MoE vision-language models with unified multimodal reasoning and 256K context. * Gemma 4 26B-A4B and Gemma 31B IT - Google's vision-language models with thinking mode and 256K context. * Autoresearch - Let Claude Code run ML experiments for you, now in a template. New guides. Other improvements. This update also brings fleet-wide mitigations for the recently disclosed CopyFail exploit. No action is required from users, and running workloads remain unaffected. Its commitment. As other clouds are out of GPU capacity or charging a premium for what's left, Vast.ai Inc. remain focused on making high-performance AI infrastructure available, affordable, and easy to deploy at scale. Need help? Contact Vast.ai Inc. anytime at [email protected] or join its Discord server for tips, community support, and platform updates in real time. Change log. New features. * Benchmarking CLI for GPU performance and cost comparison. * All-in-One App Studio: eight creative AI tools in one GPU instance. * Authenticator app support: use any TOTP app, including Google Authenticator, 1Password, and Microsoft Authenticator, instead of SMS. * CLI 2FA: 2FA now works through the Vast CLI. * Backup codes: automatically generate one-time recovery codes so you're never locked out. * Multiple authentication methods: register more than one per account. * 2FA for team roles and API key permission groups. Issues resolved. * CopyFail exploit: fleet-wide mitigations rolled out with no impact on running workloads and no user action required. * SSH key input validation, error message, and link fixes. * Cloud copy fixes. * Mobile UI fixes. * Notifications fixes. * General proactive security fixes. API changes. * The Vast pip package has been upgraded, and the SDK and CLI repos have been merged into a single, easy-to-grab repo. Use: pip install vastai. * Install its agent skill to help Claude, Codex, etc., drive its CLI: npx skills add vast-ai/vast-cli. Updated templates. * Wan2GP - migrated from provisioning installer to Docker. * Whisper UI - migrated from provisioning to Docker image. * FluxGym - Dockerized to prevent provisioning failures. * vLLM - updated build actions for new CU130 image tag.
Vast.ai named among fastest growing vendors by ramp and brex. March 5, 2026 By Team Vast