Harper

10/08/2026

Vector search just got a major upgrade in Harper 5.3. 🚀

⚡ Up to 6.5× higher throughput vs. Harper 5.2
⚡ 1.6–2.3× higher throughput vs. PostgreSQL + pgvector in our 200K-vector benchmark

Native Rust HNSW. No separate vector DB required.

https://www.harper.fast/resources/native-vector-search-in-harper-5-3

09/09/2026

9/9/26 → localhost:9926

Coincidence? Technically, yes.

But it’s also Harper Day, our extremely legitimate, definitely-not-made-up holiday.

This is your sign to build something with Harper.

https://github.com/HarperFast/harper

08/28/2026

We just published a new reference project for developers curious about real-time apps: the MQTT Live Scoring App, built on Harper.

Here's the short version of what makes it interesting. Score events publish over MQTT to Harper's built-in broker, no separate message queue to install or maintain. The event ingest logic runs in the same process as the database, so the timing numbers you see in the app's telemetry strip are real, measured server-side, not simulated for the demo. On top of that, the two data tables (matches and score events) get full REST APIs automatically from their schema definitions, no handler code required.

It's a compact project but it's a genuine look at what you can simplify when your database, broker, and API layer live in one platform instead of three.

Take a look: www.harper.fast/resources/the-topic-is-the-record-real-time-fan-out-without-a-broker-a-bridge-and-a-cache

08/25/2026

Every network hop in your architecture has a cost: latency, complexity, failure modes, and dollars.

Harper takes a different approach: put the database, cache, messaging, and application runtime in one process to remove the distance.

https://www.harper.fast/resources/the-distance-problem-why-centralized-data-paths-cost-you-latency-uptime-and-money

08/10/2026

🎉 Gartner® recognized Harper as an Honorable Mention in the 2026 Magic Quadrant™ for Cloud-Native Application Platforms

👉

Harper’s inclusion highlights a broader shift toward integrated cloud-native platforms that unify data, caching, messaging, and application logic. The article explains how Harper’s distributed runtime and managed infrastructure reduce operational complexity, improve performance, and support mode...

08/06/2026

🚀 Harper 5.2 is available now!

Get 5–8× faster warm reads, more predictable performance under heavy writes, and more throughput per node—with fewer systems to manage.

See what’s new: https://www.harper.fast/resources/harper-5-2-more-throughput-per-node-fewer-systems-around-it

07/14/2026

Web personalization is a platform decision.

Architecture determines data freshness, latency, operational complexity, crawler visibility, and cost at scale.

We compared five common approaches and the tradeoffs behind each:

https://www.harper.fast/resources/five-architectures-for-web-personalization

06/15/2026

Harper has been named a Sample Vendor in the Gartner® Hype Cycle™ for Digital Commerce, 2026 in the Front-End Cloud category.

We believe the next evolution of front-end cloud is about what the runtime can do.

Harper was named a 2026 Gartner Hype Cycle for Digital Commerce Sample Vendor, recognized in Front-End Cloud for its unified runtime combining data, application logic, messaging, and agent development together.

05/15/2026

The citizen developer era is here, but the real question is whether AI-built apps can actually run inside the enterprise.

Stephen Goldberg explains why Harper’s unified runtime, built-in governance, and deployment flexibility make it the right foundation for what comes next.

https://www.harper.fast/resources/why-harper-is-the-definitive-platform-for-enterprise-citizen-developers

Harper bridges the gap between business agility and IT security. Utilizing a unified runtime, Harper Fabric guarantees data sovereignty across any environment, from public clouds to air-gapped facilities. Empower users with secure, compliant AI application development and robust governance.

05/14/2026

Everyone is building agents on Harper right now.

And the more people build, the more compelling the case gets.

Jeff Darnton’s latest article shows one strategy for turning coding agents into a real development system: dispatched agent teams that plan, build, QA, and review in parallel, with humans still owning direction.

The big unlock: Harper’s unified runtime gives agents fewer systems to understand and more room to build.

Harper enables production-grade agentic engineering by collapsing database, cache, runtime, and messaging into one process, reducing agent complexity and review burden. A multi-model dispatch workflow lets specialized agents plan, code, QA, and review in parallel while humans retain control over cri...