💫 5,000 GitHub stars and counting! Hatchet keeps growing! Shoutout to the Hatchet team 🔥 openalternative.co/hatchet

Huge congrats to the Mergent team for joining forces with Resend to scale their email infrastructure 🚀 For anyone running tasks on Mergent: here's a migration guide to easily migrate from Mergent -> Hatchet: docs.hatchet.run/blog/mergent-m…

🚀 We built a deep research agent in Typescript in only a day -- all without an AI framework. It's able to generate a research report in < 2 min using Hatchet's built-in parallelism and durable execution primitives. And the best part: there's no lock-in. Instead of using a

Recently wrote an article about how to optimize Postgres inserts using Go, pgx and pgxpool -- here are the main takeaways:

🪓 Product Update 🪓 When you're building AI agents, it's common to allow users to create and register their own automations on your platform. We just launched a set of features which help with just that: - Event filtering -- you can dynamically register filters on your tasks

We also put together a post on avoiding blocked event loops in Python: docs.hatchet.run/blog/warning-e…

Because we're open source, we often get bundled into the on-prem deployments of our customer's customers -- which is a underappreciated network effect of OSS companies. This is becoming even more true as it's quickly becoming a standard for startups to offer a

1/14 Hatchet Documentation explores why Go is a good fit for building agents. The article delves into the characteristics of agents and how Go's features align with these requirements. Read more here: docs.hatchet.run/blog/go-agents Hatchet via X Composer ⬇️

Some of our users are shipping so quickly that their code occasionally has high failure rates. I like to call these "Christmas tree deployments" because the graphs are all red and green. Luckily, Hatchet provides the durability to ensure that Christmas tree deployments lead to

For AI agents, massive parallelization is a huge bottleneck - this startup cut processing times for processing 50k documents from > 1 week to less than an hour with Hatchet Aevy (aevy.io) processes massive volumes of documents for critical infrastructure

Startups are about focus. For the past year, the only thing that's mattered to us Hatchet is helping our users scale by optimizing throughput and latency for all possible workloads. We're now processing thousands of tasks/second for a handful of users, and our average

Jake Checkout Hatchet by Alexander Belanger

Execution engine? Powered by Hatchet . This means: 🔮 No monolithic loops 🔮 Parallel execution 🔮 Independent workers 🔮 Crash-proof flows 🔮 Massive scalability Web3 infra, made more magical.

the only primitives that should be shared across most AI agents: durable execution, memory, and observability. the argument is simple: the best AI apps aren't using frameworks. don't let artificial constraints or abstractions built by someone else impact your product direction.

we use Fly.io to power our managed compute offering and automatic scaling based on queue depth across 30+ regions. we're happy customers!

🪓 🚀 We're excited to announce a (long awaited) feature that's going to unlock some incredible use cases: real-time streaming is now live on v1 for Python, Typescript and Go! Tasks can now stream data back to consumers in real-time using the put_stream() method. This opens up

🚀 Excited to announce our next monthly AI Builders Meetup happening Sept 9 at the GitHub office in SF! We’ve got great speaker lineup from our recent call for proposals: Mira Sharma – Product Okta Patrick Lundquist – Staff Engineer Clarifai gabe ruttner – Co-founder & CTO

Created a simple quickstart for self-hosted Hatchet Python workflows w docker compose specific to Coolify github.com/tamasbelinszky…