Winter 2027
Machine Learning Software Engineer Intern
Winter 2027
Updated on 8/10/2026
Rippling
5,001-10,000 employees
Unified HR and IT management platform
No salary listed
San Francisco, CA, USA
Hybrid
Interns will be based in the San Francisco or New York office; office work is currently required at least three days per week.
Master's, PhD
 AI & Machine Learning (1) |
|---|
 Software Engineering (1) |
|---|
Get referred to Rippling
See people who can refer or advise you
- Currently enrolled in a Master of Science or Ph.D. program in computer science or a related field during the internship.
- Have solid programming skills with an emphasis on backend experience and knowledge.
- Have knowledge of Python, PySpark, and PyTorch.
- Communicate effectively.
- Translate product needs into mathematical formulations, train the resulting models, and ship them to production with a small team.
- Design scalable machine learning pipelines for data preprocessing, feature engineering, model training, and evaluation.
- Work with data engineers to collect and preprocess datasets for model training.
- Stay up-to-date with the latest research in machine learning and related fields and apply this knowledge to improve products.
- Experience developing applications that use large language models and familiarity with pre-training and fine-tuning techniques.
Rippling provides a unified SaaS platform that combines HR and IT management. It automates payroll, benefits administration, employee data management, and app/device provisioning, all within one system. The platform integrates these functions so businesses can handle HR and IT tasks from a single interface, with automation and connections to other business apps. Its approach centers on offering a subscription-based service that includes core HR tools plus optional services like device management and broker partnerships. Rippling aims to reduce administrative overhead and improve operational efficiency by keeping HR data, payroll, benefits, and IT management in one place.
Company Size
5,001-10,000
Company Stage
Series G
Total Funding
$1.8B
Headquarters
San Francisco, California
Founded
2016
Get referred to Rippling
See people who can refer or advise you
Simplify's Take
What believers are saying
- June 2026 Data Cloud already serves 560 companies and adds $5-$7 million monthly.
- Rippling’s 2025 Series G raised $450 million, valuing it at $16.8 billion.
- Rippling passed $1 billion ARR by 2025 and is still hiring aggressively.
What critics are saying
- Runlayer sued Rippling on July 28, 2026, alleging trade-secret theft and product cloning.
- Deel’s 2026 counterclaims and Rippling’s RICO trial keep executives in expensive discovery.
- If courts restrict Rippling’s AI launches, the company loses its fastest product-growth engine.
What makes Rippling unique
- Rippling Data Cloud launched June 2026, bundling HR, finance, and AI data workflows.
- Preferred Carrier Program with Guardian, MetLife, and Unum automates benefits data exchange.
- Same-day payroll banking and SOC 2 compliance tools deepen Rippling’s operating-system moat.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Hybrid Work Options
Growth & Insights and Company News
6 month growth
↑ 1%1 year growth
↑ 1%2 year growth
↑ 1%AI security firm accuses Rippling of trade secret theft. 5h ago · 0:00 listen · Source: IAM Media Summary. An AI security firm is accusing Rippling of trade secret theft. Runlayer alleges a Rippling employee revealed the company is using Runlayer's trade secrets. This information was reportedly used to build a clone of Runlayer's product. That product is software designed to control how AI agents interact with enterprise systems. This story highlights potential intellectual property disputes in the rapidly evolving AI industry. This is an AI-generated audio summary. Always check the original source for complete reporting.
Runlayer sues Rippling over alleged MCP gateway theft. Published · Jul 29, 2026 Why Relve is watching this The suit is a live test of how much access enterprise customers can get during a product trial before "build it ourselves" crosses into trade secret theft, a risk every AI infrastructure startup selling to tech-savvy buyers now has to price in. Key Takeaways * Runlayer, an MCP security gateway startup, is suing HR software company Rippling over trade secret misappropriation and breach of contract claims. * Runlayer alleges Rippling used nearly a year of deep product trial access, including its source code, to build a competing MCP gateway in-house. * A "Rippling insider" reportedly texted Runlayer's CEO to warn that Rippling's internal project was nearly a one-to-one copy of Runlayer's product. * Rippling confirmed it is launching its own MCP gateway, but its spokesperson denies misusing Runlayer's intellectual property. * Runlayer has raised $42 million since launching in mid-2025, backed by Khosla Ventures and Felicis. What happened. MCP security gateway startup Runlayer has sued HR software company Rippling, alleging trade secret misappropriation, unfair competition, and breach of contract, according to a complaint seen by TechCrunch. Runlayer says Rippling evaluated its product for nearly a year under a mutual non-disclosure agreement and a trial agreement barring Rippling from copying its technology or building derivative works. The two companies could not agree on a price, and Runlayer ended the trial. Runlayer alleges a "Rippling insider" later texted its CEO, Andrew Berman, to warn him about an internal project building a near-identical copy of Runlayer's gateway. Rippling has confirmed to TechCrunch that it is launching its own MCP gateway, while denying it misused Runlayer's intellectual property. Runlayer, which launched in mid-2025 and has raised $42 million from investors including Khosla Ventures and Felicis, has retained law firm Sullivan & Cromwell for the case. Why it matters. The dispute is a cautionary tale for AI infrastructure startups selling into enterprise customers that have their own engineering muscle, since a deep, months-long product trial can double as a blueprint for a competing in-house build. MCP gateways have gotten more crowded since Anthropic open-sourced the underlying protocol in 2024, raising the stakes for smaller vendors trying to lock in enterprise deals before customers build their own version. Runlayer's claims come from its own complaint and remain unproven, and building a similar product after a failed deal is not automatically theft, since ideas and general product categories are not protected the way specific code or trade secrets are. Retaining a marquee law firm signals confidence, not the merits of the case. Runlayer's panicked effort to avoid competition by fabricating claims. Rippling spokesperson Bottom line. Watch how this suit proceeds, since a ruling either way will shape how AI infrastructure startups structure enterprise trials and NDAs going forward. A win for Runlayer could make companies more cautious about extensive hands-on evaluations with competitors-in-waiting, while a win for Rippling would reinforce that enterprises can walk away and build their own version. For startups selling AI infrastructure into enterprise, per Relve, an AI tools intelligence platform, the practical takeaway is to limit source code and roadmap access during trials, and to price and structure NDAs with the possibility of a walkaway build in mind. Neelam Khan. Lead Editor Neelam Khan is a Lead Editor at Relve, covering AI news, tools, product updates, search trends, and business use cases. She filters noise from useful signals for founders and teams, drawing on her previous work in AI SEO, content strategy, and tool research with Wellows and AllAboutAI.
Runlayer sues Rippling over alleged copy of MCP gateway. 41 minutes ago A year-long product trial turned contentious after Runlayer says Rippling copied its MCP gateway, raising fresh risks for AI infrastructure startups. Runlayer, a startup that offers a secure gateway for the Model Context Protocol (MCP) - a standard for the safe use of external data and tools by AI models and agents - has filed a lawsuit against the HR software startup Rippling, according to TechCrunch, citing the complaint. This case serves as a warning to companies selling AI infrastructure to enterprise clients, including other tech companies that are increasingly able to develop similar solutions in-house. In the lawsuit, Runlayer describes a lengthy product evaluation Rippling conducted as a potential client: during the assessment, Runlayer shared the product roadmap and source code. The parties signed a mutual nondisclosure agreement, and Rippling signed a trial agreement with a prohibition on copying Runlayer's intellectual property or creating derivative works. According to the complaint, the evaluation lasted 'almost a year of intensive engineering collaboration.' However, the parties did not agree on price, after which the trial period was terminated. Lawsuit Context and the Parties' Positions an in-house project that effectively clones Runlayer... this is almost a complete copy of Runlayer - Runlayer's lawsuit Runlayer contends that the complaint materials point to trade secret misappropriation, unfair competition, and breach of contract. Rippling confirmed to TechCrunch that it is launching its own MCP gateway, but a spokesperson denied the accusations of using Runlayer's intellectual property. Runlayer's panicked moves to avoid competition by fabricating claims are not an effective way to solve their business problems. Rippling is launching a more sophisticated product to connect AI tools to business data, using only its own information - Newshunt has every reason to win in this market - Rippling spokesperson, TechCrunch Runlayer has enlisted the prestigious law firm Sullivan & Cromwell. While that adds some heft to the lawsuit, it does not guarantee victory in court. In the broader MCP gateway market context, competition is increasing: Anthropic launched MCP as an open protocol in November 2024, and it is now one of the foundational building blocks of AI interoperability, enabling models and agents to safely interact with external data and services. MCP gateway products add control and security, particularly for managing agents, and the sector has become much more competitive following Runlayer's launch last year and the $42 million investment from Khosla Ventures and Felicis. Even after intensive testing, an enterprise may decide to build the tool in-house. Consequently, the market underscores the difficulties of selling complex AI infrastructure in corporate environments.
Runlayer files suit against Rippling alleging trade secrets misappropriation in SDNY over AI product 'clone'; seeks preliminary injunction. Jul 28, 2026, 16:20 ET Rippling has been preparing to launch a competing product built on Runlayer's technology, text messages from Rippling engineer show NEW YORK, July 28, 2026 /PRNewswire/ - Today in the Southern District of New York, Runlayer, the platform helping companies become AI-native through the securely managed deployment of AI agents, filed a complaint against Rippling, an HR and workforce-management software company, alleging trade secret misappropriation, unfair competition, and breach of contract. Runlayer also seeks a preliminary injunction and expedited discovery. The complaint alleges that Rippling misappropriated Runlayer's trade secrets and violated confidentiality agreements during the course of a nearly year-long commercial relationship between the two companies and is preparing to launch a competing product built on Runlayer's technology. Following a product trial between the companies, an insider at the $16.8 billion Rippling texted Andrew Berman, CEO of Runlayer, that "There's been a project internally [at Rippling] to build essentially a clone" of Runlayer that is "almost a 1 to 1 copy of Runlayer." "Runlayer invests heavily in its innovations and proprietary technologies and will vigorously defend its intellectual property," said Mr. Berman. "Our platform is built on two equally critical pillars, AI enablement and control, that together help our customers safely and fully adopt AI agents. We work with customers and partners across the world in an environment of mutual trust, and just as we protect our customers, we have no choice but to ensure that competitors like Rippling cannot breach our confidentiality agreements or misappropriate our trade secrets." Runlayer has secured $42M in funding, with backing from investors including Khosla Ventures and Felicis. In just eight months since emerging from stealth, the company has become a market leader in helping enterprises adopt AI securely. Runlayer is represented by Sullivan & Cromwell LLP. SOURCE Runlayer
Runlayer sues Rippling over alleged AI gateway IP theft. technology TrendPulse AI Analysis This article covers technology trends, sourced from TechCrunch. Its AI system has analyzed the key points and extracted the most relevant insights for decision-makers. Below is the structured breakdown of the original content. Quick summary. * Runlayer has filed a lawsuit against Rippling, alleging the HR software giant misappropriated trade secrets to build a clone of its Model Context Protocol (MCP) gateway. * The legal dispute highlights the growing risks for AI infrastructure startups when sharing proprietary source code and roadmaps during enterprise product trials. * While Rippling denies the claims, the case underscores the tension between enterprise software procurement and the internal development capabilities of tech-native companies. Key details. Runlayer, a startup specializing in secure MCP gateways, claims that Rippling utilized an extensive year-long product trial to gain unauthorized access to its intellectual property. According to the lawsuit, the two companies engaged in deep engineering collaboration under a mutual non-disclosure agreement, during which Runlayer shared sensitive product roadmaps and source code. Following a breakdown in pricing negotiations, Runlayer alleges that Rippling pivoted to developing an internal, near-identical clone of the platform. Legal counsel for Runlayer, the firm Sullivan & Cromwell, argues that Rippling's actions constitute trade secret misappropriation and breach of contract. Conversely, a spokesperson for Rippling has dismissed the allegations as a "panicked effort" by a struggling competitor, asserting that their internal development relies solely on proprietary information. The company maintains that its forthcoming product is a superior solution for connecting AI agents to business data. Runlayer's panicked effort to avoid competition by fabricating claims is not an effective way to deal with its business failures. Rippling is launching a superior product for connecting AI tools to business data using only its proprietary information - TrendPulse has every reason to win in this market. Why this matters. This case serves as a critical warning for B2B AI startups navigating the enterprise sales cycle. As Anthropic 's open-source MCP standard gains traction, the market for gateway security and management tools is becoming increasingly crowded. Startups are often forced to provide deep technical access to potential enterprise clients to prove their value, creating a vulnerability where large, well-funded tech companies can effectively "audit" a startup's innovation before deciding to build it in-house. For investors and executives, this dispute illustrates the "build vs. buy" dilemma that defines modern AI infrastructure. When a customer possesses the engineering talent to replicate a niche tool, the barrier to entry for startups becomes razor-thin. Moving forward, startups must balance the necessity of transparent trials with rigorous IP protection strategies to ensure their technical differentiation remains defensible against larger incumbents. The bottom line. The Runlayer-Rippling lawsuit highlights the high-stakes risks of enterprise AI trials, where deep technical collaboration can inadvertently provide incumbents with the blueprint to displace their own vendors. This article has been processed and analyzed by TrendPulse AI for informational purposes. Content may have been summarized or restructured for clarity.