Inside AI Engineer Paris 2025 Part 1 – 5 Highlights That Shaped the Stage
Last week, AI Engineer Paris took place. With 5 talk tracks and 24 sponsor booths, there was a ton of content for AI engineers and ML practitioners. Find out what the themes and trends were for the future of AI development.
Scale-to-Zero: Wake VMs in 200ms with Light Sleep, eBPF, and Snapshots
Discover how we built Scale-to-Zero to automatically wake VMs in 200ms with eBPF, light sleep, and snapshots.
app.build by Neon: Spawning 1000s of AI-generated apps at scale with Koyeb
Learn how app.build, Neon’s open-source AI agent, uses Koyeb to deploy thousands of full-stack apps generated by AI—automatically and at scale.
How eXalt Built a Secure and Scalable ChatGPT Alternative with Koyeb
Learn how eXalt built a secure, scalable ChatGPT alternative using Ollama to serve models like Mistral, Gemini, and DeepSeek, all running on Koyeb’s serverless GPUs.
How Anyshift Scales Real-Time Queries Across Millions of Nodes with Koyeb
Discover how Anyshift, an AI-powered infrastructure assistant, leverages Koyeb’s serverless platform to scale, prototype faster, and manage high-performance infrastructure with ease.
Achieve 5x Faster Inference Speeds on Serverless GPUs with Pruna AI and Koyeb
Today, we are excited to announce our partnership with Pruna AI. Pruna AI is the optimization engine built to simplify and accelerate scalable inference. Koyeb offers a serverless cloud platform for teams to deploy ML and AI models on high-performance GPUs, CPUs, and accelerators - globally.
Building Models 10X Faster: Why Scenario Chose Koyeb
Discover how Scenario optimizes their models 10x faster with Koyeb. Seamless GPU switching, high-speed storage, and ready-to-use dev environments lets Scenario focus on building AI models, not heavylifting infrastructure.
A2A and MCP: Start of the AI Agent Protocol Wars?
Google just released A2A, a new open protocol for agent to agent communication, and it left people wondering if they were starting an agent protocol war against MCP.
Best Serverless GPU Platforms for AI Apps and Inference in 2025
Explore our comparison of the leading serverless GPU providers and compare prices across H100, A100, and L40S serverless GPUs.
Orbit Codes: Achieving 10x Faster Deployments and Global Scale with Koyeb
Discover how Orbit Codes optimized deployments, achieved seamless global scaling, and reduced costs by switching to Koyeb.
Best Open Source Multimodal Vision Models in 2025
Discover top multimodal vision models in 2025: Gemma 3, Qwen 2.5 VL 72B Instruct, Pixtral, Phi 4 Multimodal, Deepseek Janus Pro, and more. Deploy on serverless GPUs for scalable, dedicated inference endpoints.
eToro Accelerates Deployments for Real-Time Apps with Koyeb
Discover how eToro streamlined their development process for real-time applications with high-performance managed infrastructure, built-in continuous deployments, and autoscaling.
Globula: Autoscaling seamlessly to 10,000 players and beyond
Discover how Globula, a geo-location based AR game, scales to thousands of players using autoscaling. Learn how they accelerate and optimize their deployments with Scale-to-Zero, continuous deployment, and more.
Best Open Source LLMs in 2025
Explore top open LLMs in 2025: DeepSeek-R1-Qwen-32b, Mistral Small 3, and Qwen 2.5 Coder. Deploy on serverless GPUs for scalable, dedicated inference endpoints.
Deploy AI Infrastructure in 2025: Serverless GPUs, Autoscaling, Scale-to-Zero, and More!
Discover how high-performance serverless GPUs, autoscaling, scale-to-zero, and other cutting-edge features simplify AI deployment in 2025. Learn how our next-gen platform empowers developers to build faster, smarter, and more cost-efficient AI solutions.
Top PostgreSQL Database Free Tiers in 2025
Top PostgreSQL providers with free hosting solutions. Discover free options ideal for building side projects and MVPs.
December Recap: Scale-to-Zero, Serverless GPU Price Drop, and more
Since we've released so much goodness in the past couple of weeks, we prepared a recap to make sure you don't miss out on any of our serverless news.
Pro and Scale Plans: Manage 1000s of Services, More Users, and Included Compute
Discover our new plans designed for developers and teams to provide more flexibility, clearer pricing, and effortless scalability as your needs grow.
Serverless GPUs: Slashing L4, L40S, A100 Prices and Increasing Efficiency
We are dropping prices across our range of Serverless GPUs including L4, L40S, and A100 GPUs. Build and deploy your AI applications for less with serverless using Scale-to-Zero and Autoscaling.
Beavr Chose Koyeb for a Next-Gen Heroku and Vercel for the Backend Experience
Discover how Beavr scaled their SaaS platform globally with Koyeb and why they chose Koyeb infrastructure over Vercel for scalability, autoscaling, and enhanced developer experience.
Autoscaling GA: Scale Fast, Sleep Well, Don't Break the Bank
Today, we are announcing autoscaling in GA! Get dynamic, flexible, and responsive resource allocations for production. Automatically scale your AI and full stack applications.
Best LLM Inference Engines and Servers to Deploy LLMs in Production
Looking to boost the performance of your AI workloads using LLMs in productions? Explore the best inference engines and servers like vLLM, RayLLM with RayServe, TensorRT-LLM, HuggingFace Text Generation Inference, and more to see which one you should be using when performing inference.
The engineering behind autoscaling with HashiCorp's Nomad on a global serverless platform
Wonder how we added the ability to automatically scale Services running on our platform? Learn about the engineering behind autoscaling Instances on a global and serverless platform.
What are LLMs? An intro into AI, models, tokens, parameters, weights, quantization and more
Trying to keep up with AI and all the buzzwords surrounding it? Learn some essential AI terminology like parameters, weights, tokens, quantization, sparsity and more with our intro to LLMs.
Meet Paweł, Software Engineer Building the Koyeb Serverless Platform
The Koyeb team is growing! Get to know Paweł, software engineer building the Koyeb serverless platform.
toddl.co: Spain's Leading Platform for Extra-Curricular Activities Deploys 10x Faster with Koyeb
Learn how toddl.co reduced its build and deployment times, automated manual proccesses, and regained time to focus on its business.
Ollama and Friends' Local and Open Source AI Developer Meetup at KubeCon Paris
Last Thursday night, we co-organized a Local and Open Source AI developer meetup with Ollama and Dagger at Station F. Over 450 developers attended, both from the local and international scene.
KubeCon Paris Events to Attend 2024
KubeCon EU 2024 is upon us! We are gearing up for a week full of cloud native and open source fun! Check out our handcrafted list of this week's awesome events.
What is RAG? Retrieval-Augmented Generation for AI
Learn all about RAG, the AI framework addressing major LLM limitations by supplementing an LLM's knowledge source with external resources. Discover the benefits of RAG, its origins and ideal use cases, how to implement it and deploy RAG-powered AI applications on Koyeb.
Deploy Apps and Containers in Singapore on High-Performance Infrastructure GA
We are thrilled to announce that our Singapore location is generally available to deploy your full stack applications, low-latency AI workloads, APIs, and databases.
FoxSell's Journey to Time, Cost, and Performance Optimization with Koyeb
Discover how FoxSell optimized time, cost, and performance with Koyeb's innovative cloud solutions. From streamlined global deployments to exceptional performance on bare metal machines, explore their journey to efficiency in e-commerce.
University of British Columbia HART Accelerates Deployments with Koyeb
Discover how the University of British Columbia Housing Assessment Resource Tools (UBC HART) leverages Koyeb for rapid deployments of the housing assessment resource tools they are developing in the face of Canada's housing crisis.
Meet David, Product Manager orchestrating all the work on the Koyeb Serverless Engine
Our team is growing! Get to know David, Koyeb's first Product Manager, and find out how he is helping us make our vision for serverless a reality.
Meet Leo, Software Engineer building the Koyeb Serverless Engine
Our team is growing! Get to know Leo, Koyeb's newest Software Engineer, and find out how he is helping us make our vision for serverless a reality.
Kong Konnect the World: Seamless Global and Serverless Deployments Powered by Koyeb
Kong leverages Koyeb for rapid global deployments to highlight its API lifecycle management platform.
Qwigo Leverages Koyeb for Rapid Global Deployments and High-Performance MicroVMs
Discover how Qwigo leverages global serverless deployments on Koyeb to deploy their CPU-intensive services across the US and Europe.
Which Cloud Database Platform to Choose for Your Applications
Using a managed database saves time and every when building applications that need persistent storage. While there are numerous benefits to using a managed database, there are serious and long-term considerations to have before choosing one for your app. Our analysis aims to equip you with insights that will help when selecting a managed database for your next project.
Sustaining free compute in a hostile environment
Today, we reaffirm our commitment to maintaining a free tier. We explain how we intend to sustain it and why we are so committed to providing one. We provide a free tier so users can explore the platform and deploy a hobby project before deploying a production-grade application. This is a story about bare metal, abuse, credit cards, and scale-to-zero.
Building a global deployment platform is hard, here is why
Deep dive into how we built a global serverless engine with Nomad, Kuma/Envoy and Golang to make multi-region deployments easy.
Koyeb Metrics: Built-in Observability to Monitor Your Apps Performances
With Koyeb Metrics, you get a high-level overview of what is happening in your services running on Koyeb. Discover how you can use Koyeb Metrics to understand your services and diagnose performance issues.
Dockerfile Deployment on High-Performance MicroVMs is GA
Today, we are excited to announce the support of Dockerfile based deployments in general availability. Building and deploying using Dockerfiles offer more flexibility and control over the build process of your applications to let you deploy any kind of applications, frameworks, and runtimes.
Deploy and scale high-performance background jobs with Koyeb Workers
Today, we are thrilled to announce workers are generally available on Koyeb! You can now easily deploy workers to process any background jobs with high-end performance in all of our locations.
Koyeb CLI 3.0: Better flows, improved troubleshooting, and reworked foundations
The latest version of the Koyeb CLI is available and brings many helpful improvements! Discover the reworked error messages, smoother flow for creating and updating services, and how we improved the foundations of our CLI to continue to building the developer experience we envision.
Meet Justin, Technical Writer and Documentation Expert
The Koyeb team is growing! Get to know Justin, our technical writer and driving-force behind our documentation and technical guides.
Meet Sebastian, our Customer Success Engineer ensuring seamless deployments
Exciting news: We have a new teammate! Get to know Sebastian, our Customer Success Engineer helping Koyeb users deploy successfully!
What is a microVM?
Want to learn more about microVMs? Learn all about this lightweight virtualization technology and get all of your microVM questions answered with this short and sweet primer.
Meet Kamil, Product Designer optimizing DX
Get to know our new team member, Kamil! He is our product designer helping us build an optimal developer experience and make our vision for serverless a reality.
What is gRPC?
Learn all about gRPC, a high-performance remote procedure call framework, and how it improves communication between services. Discover the benefits of using gRPC, its use cases, how to implement it in various programming languages and deploy gRPC applications on Koyeb.
Meet Julia, Talent Partner building our dream team
We have a new team member! Get to know Julia, our talent partner helping us make Koyeb's vision for serverless a reality. Learn what she looks for in a candidate, her favorite part of the hiring process, and what she gets up to outside of work.
eBPF: The future of the service mesh and network innovation
eBPF lets you run sandboxed programs in a kernel's operating system. Catch up on the heated debate taking place in the service mesh world about how this technology will shape the future of the service mesh and network innovation.
Meet Julien, Software Engineer building the Koyeb Serverless Engine
We have a new team member! Get to know Julien, a Software Engineer helping us make Koyeb's vision for serverless a reality.
US-East region is live: deploy your apps in Washington, DC
You can now deploy your full-stack applications and APIs in our newest US-East location near Washington, DC. Enjoy high-end performance and all of the platform's built-in features for your apps in the world's largest connectivity hub.
New Frankfurt location: deploy high-performance apps in Europe
Today, we are super excited to announce the grand opening of our Frankfurt core location in Germany! Frankfurt is the largest connectivity hub in Europe and is an amazing place to run high-performance and low latency applications.
What is continuous deployment?
Learn about continuous deployment, what goes into building a strong continuous deployment pipeline, the value it adds, how it differs from continuous delivery, and getting it built into your application development.
What is a service mesh?
Wondering what is a service mesh? Get all of your service mesh questions answered with this short and sweet primer. We take a look at how the two crucial parts of a service mesh, the control plane and the data plane, work together to handle interservice communication.
Distributed tracing with Envoy, Kuma, Grafana Agent, and Jaeger
Discover how we added end-to-end tracing to all requests for Koyeb Apps. We explain how we implemented end-to-end tracing, why we chose Jaeger and Grafana Agent to power our observability stack, and how we overcame the challenges we encountered along the way.
Heroku’s free tier legacy: The shoulders we stand on 15 years later
Heroku's free tier changed the way developers, hobbyists, students, and indie hackers deployed applications. Heroku's announcement to sunset their free tier marks the end of an era. We take the time to reflect on the rise and impact of Heroku’s legendary free tier.
What is an API Gateway?
Sitting between clients and backend services, API gateways have a number of uses and benefits. Get the lowdown with this short and simple post on API gateways.
Meet Nils, Full Stack Engineer building the Koyeb Web Console
Our team is growing! Get to know Nils, our first Full Stack Engineer, and find out how he is helping us make our vision for serverless a reality.
Meet Anthony, Senior Software Engineer building the Koyeb Serverless Engine
We have a new team member! Get to know Anthony, a Senior Software Engineer helping us make Koyeb's vision for serverless a reality.
Meet Diego, Software Engineering Intern building the Koyeb Observability Pipeline
Koyeb has a new team member! Get to know Diego, our Software Engineering Intern, and find out how he is helping us make our vision for serverless a reality.
The true cost of Kubernetes: People, Time and Productivity
While writing a comparison of Kubernetes and Koyeb, we tried to determine how much operating a Kubernetes cluster really costs. This section of our comparison took us hours to write and ended up being so long that we decided to write a dedicated post about it. Kubernetes is a proven technology, but the true cost is often underestimated: this post investigates the actual financial costs of using Kubernetes.
Blue-Green, Rolling, and Canary: Continuous Deployments Explained
If you're afraid to push to production on a Friday, rely on big-bang deployments, or find recovering from an infrastructure failure is a painful and time-consuming incident, then it is seriously time to talk about continuous deployment best practices. Discover the different go-to continuous deployment strategies and how you can get a continuous deployment pipeline built-in to your application by deploying on the Koyeb Serverless Platform.
The Team: Meet Nicolas, Senior Backend Engineer building the Koyeb Serverless Engine
The Koyeb team is growing! Get to know our new team member, Nicolas, in this interview. He is a Senior Backend Engineer helping make Koyeb's vision for serverless a reality.
Building a Multi-Region Service Mesh with Kuma/Envoy, Anycast BGP, and mTLS
We recently wrote about how the Koyeb Serverless Engine runs microVMS to host your Services but we skipped a big subject: Global Networking. This is a deep dive to understand the life of an end user's request for a service hosted on Koyeb. We explore the technology and components that make up our internal architecture by following the journey of a request from an end-user, through Koyeb's Global Edge Network, and to the application running in one of our Core locations.
The Team: Meet Thomas, Senior Backend Engineer building the Koyeb Serverless Engine
The Koyeb team is growing! Get to know our new team member, Thomas, in this interview. He is a Senior Backend Engineer helping make Koyeb's vision for serverless a reality.
Why you need to build globally distributed applications
Users have certain expectations for modern web services and applications. Discover how building distributed and global architectures enable you to respond to those standards and and what it really means to deploy globally..
Understanding REST, gRPC, GraphQL, and OpenAPI to build your APIs
There are several different architecture designs for Web APIs. While REST and RPC remain two popular choices, the arrival of GraphQL and OpenAPI bring new possibilities in the realm of performance, functionality, productivity of your web APIs.
The Koyeb Serverless Engine: from Kubernetes to Nomad, Firecracker, and Kuma
We decided to build our own serverless engine, one that would not be limited by existing implementations. The first version of Koyeb was built on top of Kubernetes and allowed us to quickly build a working cloud platform. After a few months of operating with this version, we decided to move user workloads from Kubernetes to a custom stack based on Nomad, Firecracker, and Kuma.
API Gateways: Improving performance, security and management of microservices
An API gateway is an API management tool that provides several benefits in a microservice architecture. Learn more about how API gateways work, their typical use cases, and what you should consider before implementing one.
Using Cache-Control and CDNs to Improve Performance and Reduce Latency
Caching is an effective technique for improving performance and reducing latency speeds for the requests of your web services and apps. CDNs bring your content even closer to end-users. Learn about cache control: what it is, how to configure it, and when to use it.
Service Mesh and Microservices: Improving Network Management and Observability
A service mesh is a dedicated layer of infrastructure that simplifies network management and increases visibility into typically complex microservice architectures. We explore this emerging technology by reviewing its history, purpose-built design, and implementations.
Lightweight Virtualization: the Container Ecosystem and Firecracker MicroVMs for Serverless
Virtualization optimizes the use of computing resources. Firecracker, a lightweight virtualization technology, is transforming the possibilities of serverless workloads.
RabbitMQ vs Apache Kafka: Comparing Message Brokers and Event Streaming Platforms
Event routers are the middlemen in an event-driven system. RabbitMQ and Apache Kafka are two popular event routers with very different implementations. Learn about their difference to make better decisions for powering modern apps.
Service Discovery: Solving the Communication Challenge in Microservice Architectures
Service discovery is the vital component in a microservice architecture that enables communication between services. Discover the influence of DNS on service discovery as well as learn about the different models of service discovery and their real-world implementations.
Introduction to Synchronous and Asynchronous Processing
Sync and async are two popular types of programming models when building event-based architectures, APIs, and handling long-running tasks. This blog posts compares async and sync processing as well as covers when to use them.
Understanding Event-Driven Architecture and Serverless Opportunities
Event-driven architectures are a great model to align your business with the real-world. Pairing it with serverless technology is a dream come true for your developers and your business.
FaaS vs CaaS: Comparing Use Cases and Responsibilities
When considering a FaaS or CaaS deployment strategy, it is worthwhile to consider the difference in the managed responsiblities between FaaS and CaaS offerings. Learn more about ideal use cases and when to use FaaS or CaaS solutions.
10 Reasons Why We Love Firecracker MicroVMs
Firecracker is a virtualization technology with that is setting the serverless world ablaze. While there are many perks with Firecracker, here are our top ten reasons why we love Firecracker.
Cloud Computing and Serverless Architectures: What are FaaS and CaaS?
FaaS and CaaS are two popular deployment strategies with their own unique advantages and ideal use cases. Knowing what distinguishes them can help when deciding how to build and deploy your web apps.
Firecracker MicroVMs: Lightweight Virtualization for Containers and Serverless Workloads
Virtualization technology is evolving. Firecracker is an emerging solution that combines the security and isolation of bare metal instances with the density and performance of containers.
Going Serverless: Implications, Benefits and Challenges
The serverless computing era is here. Learn about the implications of going serverless as well as the benefits and existing challenges to implementing this emerging technology.
From Cloud Computing to Serverless: The rise of new paradigms
The serverless computing era is here. Learn about the history and evolution of cloud computing to see why developers and businesses are excited about serverless technology.