Upstash
Upstash is a serverless data platform providing low-latency, high-scalability data services for developers building real-time applications without managing infrastructure. The company focuses on creating the best possible developer experience with a fully serverless model, per-request pricing, and automatic scaling.
Founding Story
The founders (Enes, Mehmet, and Bilal), who are also developers themselves, felt the frustration of using tools that offer a poor developer experience. They realized the importance of a robust cloud offering and a data platform that can scale and support multiple use cases. They started Upstash as a small team with the goal of creating the best possible developer experience for anyone using their tools, eliminating the need to deploy and maintain databases yourself.
Discussions
No discussions yet
Be the first to start a discussion about Upstash
Leadership
Founders
Enes Akar
CEO and Co-founder. Previously CTO at Hazelcast Cloud (2017-2020), VP of Engineering at Hazelcast (2014-2017), Co-founder at Hazelcast (2011-2014), Engineering Leader at Confluent (2020), and various engineering roles at Ericsson and Alcatel-Lucent. Over 10 years of experience in distributed systems, serverless, edge, and cloud computing. Based in Palo Alto/Silicon Valley.
Mehmet Dogan
CTO & Software Architect. Previously Distinguished Engineer at Hazelcast (2014-2020), Chief Software Architect at Hazelcast (2011-2014), and Software Engineer at Layer (2020-2021). Extensive experience in distributed systems, databases, big data, and scalability. Based in Ankara, Turkey.
Bilal
Co-founder with deep understanding of developer needs and distributed systems experience. Limited public information available about specific background.
Executive Team
Enes Akar
CEO and Co-founder
Previously CTO at Hazelcast Cloud, VP of Engineering at Hazelcast, Co-founder at Hazelcast, and Engineering Leader at Confluent. Over 10 years of experience in distributed systems and cloud computing. Based in Palo Alto, California.
Mehmet Dogan
CTO & Software Architect
Previously Distinguished Engineer at Hazelcast and Chief Software Architect at Hazelcast. Expertise in distributed systems, databases, and scalability. Based in Ankara, Turkey.
Business Model
Revenue Model
Consumption-based pricing model with pay-as-you-go per-request pricing. Users don't pay until data starts moving through the system. Also offers fixed pricing plans for predictable costs and enterprise plans for advanced needs. Revenue streams include: API usage charges (per-command/per-request), database storage fees, fixed monthly subscriptions, and enterprise contracts.
Pricing Tiers
256MB data, 500K commands/month, 10K commands/sec, 10GB bandwidth
100GB data, unlimited bandwidth, $0.25/GB storage
Multiple tiers: 250MB ($10), 1GB, 5GB, 10GB, 50GB, 100GB, 500GB ($280). No per-command pricing.
100K+ commands/sec, unlimited bandwidth, professional support with SLA, dedicated resources, HIPAA compliance
10K daily queries/updates, 200M max vectors×dimensions, 1,536 max dimensions
Unlimited daily queries, 2B max vectors×dimensions, 3,072 max dimensions, $0.25/GB storage
1M daily queries/updates, 2B max vectors×dimensions
1K messages/day, 50GB bandwidth, 1MB max message size
Unlimited messages, 50GB bandwidth, 10MB max message size
1M messages/day, 1TB bandwidth, 50MB max message size
10M messages/day, 5TB bandwidth
1K steps/day, 50GB bandwidth, 1,000 max steps per run
Unlimited steps, 50GB bandwidth
1M steps/day, 1TB bandwidth
20K monthly queries, 200K max documents
Unlimited queries, 2M max documents, $0.1/1K docs/month storage, $1/1K reranking requests
Uptime SLA, RBAC, Encryption at Rest, SOC-2, Prometheus, Datadog
Target Markets
- Developers building serverless applications
- Edge computing platforms (Cloudflare Workers, Fastly, Vercel Edge)
- AI/ML developers working with LLMs and embeddings
- Multi-region serverless architectures
- Startups and scale-ups requiring scalable data infrastructure
- Enterprise businesses with serverless needs
- Caching (general and session caching)
- Session management
- Rate limiting
- Leaderboards
- Chat applications
- Edge functions (Cloudflare Workers, Fastly Compute)
- Dub
- OpenArt
- Dropee
- Zapier
History & Milestones
Launched Upstash Vector, a serverless vector database for AI and LLMs
Reached $1M ARR and 85,000 developers; announced Series A publicly
Launched Upstash Workflow and announced deprecation of Upstash Kafka
Launched Upstash Realtime for Next.js and serverless platforms
Closed $10M Series A funding round led by a16z
1 AI Tool by Upstash
Context7
8moKeep your AI agents grounded in real, up-to-date library docs and examples using version-aware context injection.
