PIPESHIFT announced a $ 2.5 million seed round, which aims to provide enterprises with an effective construction, deployment and management of an infrastructure required for open source AI models. With more than 80 % of companies moving towards the open source Genai, the Pipeshift solution eliminates common bottlenecks related to privacy, control and engineering overhead.
Revolutionary Genai’s mlop
The new era platform of Pipeshift-AS-A-Service (PaaS) accelerates AI arrangements by providing modular MLOPS stacks running on any environment (Cloud or local). This end -to -end service covers a variety of workloads, including LLMS,,,,, Visual model,,,,, Audio modelAnd more. PIPESHIFT HANDS companies directly control their infrastructure instead of acting as GPU agents, but enable them to:
- Expand from the first day With the built -in automatic preparation, load balancer and scheduling program
- Fine -tuning or distillation Open source model with real -time tracking training indicators
- Save GPU fee Filter the model through the thermal -tone model without the GPU memory score
- Maintain enterprise -level security Maintain the exclusive data and IP all inside
By merging these functions into a single platform, the pipeline summarizes the deployment work process and greatly reduces the production time.
Powerful industry support
The $ 2.5 million seed round was led by Y Combinator and Senseai Ventures, and received additional support from Arka Venture Labs, Good News Ventures, NiveSha Ventures, Astir VC, Gradcapital and MyASIAVC. Like Kulveer Taggar (CEO of Zuess), Umur Cubukcu (CEO of Ubicloud, former leader of Azure Postgresql), and Krishna Mehra (co -founder of hat technology), etc. Essence
Co -founder and CEO of PIPESHIFT Arko Chattopadhyay,explain: “2025 is a year from Genai’s transition to production and engineering teams, and witnessed the benefits of using open source models inside. This provides high -level privacy and control and enhances performance and reduced costs. However, this is a complex and complicated and complicated and complicated and complicated and reduced cost. The expensive process involves sewing multiple components. Pipeshift’s enterprise -level arrangement platform eliminates the demand for this extensive engineering investment, which can not only simplify deployment, but also increase production throughput to the greatest extent. “
Rahul AgarwallaSenseai Ventures’s executive partner, observed: “Enterprises prefer the benefits of open source GENAI to privacy, model ownership and reduction of costs. However, transitioning Genai to production is still a complex and expensive process, and multiple components need to be sewn.” He continues: “PIPESHIFT’s enterprise -level arrangement platform can not only simplify deployment, but also increase production throughput, thereby eliminating the demand for this extensive engineering investment.”
Yash hemarjArka Venture Labs Founding partners and ordinary partners of BGV said:“We have invested in the pipeline because their innovation platform meets the key needs of the company’s AI, and seamlessly deploys the open source language model. It is deeply impressed.
Why the pipeline upgrades stand out
Founded by Arko Chattopadhyay,,,,, Enrique ferrao,,,,, PIPESHIFT’s core team Pranav Reddy has been solving the challenge of AI arrangements before this seed funding. Their experience includes an extension of more than 1,000 employees to expand an enterprise search application driven by LLAMA2. This is how tricky and resource -dense private AI deployments are introduced.
The main difference includes:
- Cloud arrangement: PIPESHIFT seamlessly handle the mixing of clouds and local GPUs to ensure cost optimization and rapid fault transfer.
- Kubernetes cluster management: The end -to -end control panel allows enterprises to create, expand and supervise the Kubernetes cluster without taking into account multiple tools.
- Model fine -tuning deployment: Engineers can use custom data sets or LLM logs for revised or distilled open source models, and can view training indicators in real time.
- 360 ° Observation: Integrated instrument board tracking performance to achieve rapid failure and effective scaling.
Designed for production, not only experiments
Companies usually use universal API -based solutions for internal privacy. PIPESHIFT flip the model by focusing on security, local or cloudy deployment. This platform supports more than 100 large language models, including Camel 3.1,,,,, MistralAnd like Deepseek encoderEssence This diversified choice helps to meet the specific performance, cost and compliance needs of each project.
Some noteworthy benefits include:
- GPU infrastructure costs as high as 60 % savings
- 30 times the production time faster
- Compared with GPT/Claude, the cost is 6 times lower
- Reduce 55 % of engineering resources
Look forward to the future
PIPESHIFT has cooperated with more than 30 companies, including NetApp, and plans to further provide a strong open source AI solution without the need for ordinary complexity. Its single -fan glass method has carried out special employment and continuous account management of MLOPS to ensure that enterprises can focus on using AI business results, rather than fighting infrastructure.
With private data protection, mixed cloud compatibility and modular flexibility, PIPESHIFT is expected to meet the various needs of enterprise AI projects. By bridging the gap between open source innovation and enterprise -level requirements, the company is an agile, security and cost -effective AI deployment in the new era.
PIPESHIFT provides end-to-end MLOPS arrangement for open source Genai workload-EmbedDings, Vector database, LLMS, VISION MODELS, and Audio Models-Abross any cloud or local GPU distribution.
