If you’ve ever been interested in up leveling your observability of your AI, particularly with respect to data, come check out how this talk (with special guest star @expanso.io :))

We're LIVE on Product Hunt! 🚀

200+ tools for AI agents at skills.expanso.io

Gives your agents tools to filter at source:
→ remove-pii
→ gdpr-enforcement
→ encrypt-data
→ and 195+ more

Self-hostable. One install command.

👉 Upvote/comment if useful to you: www.producthunt.com/products/exp...

Something drops tomorrow at midnight. 🕛

30 reusable data processing recipes for AI agents — PII removal, log aggregation, GDPR routing, encryption, and more.

One install command. Production-ready. Open source.

Set your alarm. 👀

@ProductHunt

Your AI Agents Are Running Naked Millions of API keys are being leaked through agent interactions. Here's how to secure your AI agents without sacrificing performance.

Your AI agents have raw credentials, API keys, and full file access. That's ... not good.

We built an infrastructure layer between your agents and your data. Framework-agnostic. Works with OpenClaw,@langchain.bsky.social .. you name it.

Take back control.

expanso.io/blog/2026-00...

News: We are proud to announce a new partnership between Expanso and Coginiti, which is designed to help organizations with distributed data.

If you believe "there's got to be a better way to handle our global organization's data," we can confidently say THERE IS. 😄

exso.cloud/expanso-cogi...

Three Years of Working on Global Data Three years ago, we started Expanso with a belief: data shouldn't cost more to move than to use. Here's what we've learned about upstream data control.

Hi Everyone! Took a TEENY break from blogging, and we're back! A bit of a history of Expanso, our philosophy, and what we see next :)

www.expanso.io/blog/2025-00...

Code runs. Results appear.

- No keys to manage
- No storage config to maintain
- No hunting for outputs

Bacalhau 1.8 makes storage feel like it should: invisible.

https://bac.al/180day4

Bacalhau v1.8.0 - Day 3: How Bacalhau Boosts Daemon Job Reliability A deep dive into the new orchestration logic that makes daemon jobs smarter, faster, and more resilient to infrastructure changes

Nodes come and go. Your job shouldn’t care. 

We've supercharged Bacalhau's orchestration to make daemon jobs faster and more reliable than ever. They now deploy instantly to new nodes as they join your cluster.

✅ Automatic
✅ Resilient
✅ Zero manual intervention

Bacalhau v1.8.0 - Day 2: Rerun, Update, and Version Your Bacalhau Jobs How we're making job management less of a chore and more of a superpower

Remember j-f47ac10b-58cc-4372-a567-0e02b2c3d479?

Of course not.

Bacalhau 1.8 lets you name your jobs, rerun them, version and even diff your jobs before you run them. 

Read the full breakdown on our blog to see how it works in detail →

Announcing Bacalhau v1.8.0: Intelligent Edge Computing Meets Enterprise Integration Discover how Bacalhau v1.8.0 transforms distributed computing with a native Splunk integration, name-based job management, and enhanced daemon orchestration.

We’ve made distributed computing radically more usable and radically more cost-efficient: 

 • 📊 Native Splunk integration: Slash logging costs by up to 80%
 • 🏷️ Name-based jobs: No more cryptic UUIDs
 • ⚙️Enhanced daemon reliability for services at scale 

Read the full release →

Expanso Launches Cost Optimization Platform to Cut Enterprise Data & AI Costs by up to 80% As data and AI conferences ramp up with promises of more features, more data, and more spend, Expanso is launching a different message: you should be paying ...

As Data + AI Summit rolls on—with more features, more data, and more spend—we’re focused on something else:

You should be paying less.

Expanso cuts data infrastructure costs by up to 80%, without disrupting your stack.

Try it on 10 servers. Or 10,000.
See how much you save in 60 days.

Cross-Border Data Processing with Privacy Compliance Through Bacalhau Usign Bacalhau to handle complex data pipelines that cross borders while preserving privacy

Cross-border data flows are tricky.

We wanted to see if we could build a real compliant pipeline from scratch - w/ oss tools.

✅ Generate sensitive data in the EU
✅ Anonymize it
✅ Send to US for processing

Bacalhau handles orchestration. Presidio handles privacy.

open.substack.com/pub/bacalhau...

Your Fast Track to Bacalhau: Local Development via Docker-in-Docker Docker-in-Docker With Bacalhau For Fast Tests And Low Barrier Entry

Read it here: blog.bacalhau.org/p/your-fast-...

Streamlining Bacalhau Development with the Power of Docker-in-Docker.

Our latest blog explores a practical solution using Docker Compose and Docker-in-Docker to create a self-contained, local Bacalhau environment.

Learn how to get your local Bacalhau instance running quickly and efficiently.👇

Getting Started with Machine Learning on Bacalhau Distributed Machine Learning needn't be compex with the help of Bacalhau

➡️ Read the full blog post: blog.bacalhau.org/p/getting-st...

Enjoy!

Want to set up an open-source, distributed ML pipeline that respects geographical and regulatory restrictions and runs compute in the same location as your data?

This post gets you started with Bacalhau to set up nodes in three different regions, analyze data, all while respecting data sovereignty.

High-Scale Data Processing: Over Thousands of Devices With Azure Cosmos DB and Expanso If you're dealing with data generated across thousands of devices or locations, you know the pain.

blog.bacalhau.org/p/high-scale...

How do you handle data from thousands of distributed sources before it hits a DB like Azure Cosmos DB?

We joined on Azure Cosmos DB TV to discuss exactly that!

Read the breakdown & watch the full episode with a demo in the first comment! 👇

Big news: Expanso has been selected for Plug and Play Seattle’s first startup batch!

Bacalhau is redefining what comes after the cloud:
✅ Run compute where data lives
✅ Cut latency & compliance risks
✅ Stay fast, sovereign & efficient

Honored to join this AI & infrastructure-focused cohort!

Data Breakthrough Award 2025 Winner Logo

We just won Open Source Data Platform of the Year in the 2025 Data Breakthrough Awards. 2 years, 2 wins — last year: Data Processing Solution of the Year

🔗open.substack.com/pub/bacalhau/p/expanso-w...

Huh... this looks pretty cool. Want to know what it is? Come by my talk at Kubecon at 12:10 GMT in ICC Capital Suite 1-3 tomorrow! Or check us out at the Intel booth! | Kubernetes Cross-Zone/Region Simplified: Harnessing Bacalhau for Efficient Distributed Compute

bit.ly/4bTRY5q

Promotional graphic for Bacalhau v1.7 featuring the title ‘Distributed Data Warehouse with Bacalhau and DuckDB’ on a blue background with binary code and abstract wave patterns. Logos for Bacalhau and DuckDB are displayed at the top.

New guide: Build a distributed data warehouse with Bacalhau + DuckDB.

Run SQL on regional data (EU + US) without moving it—ideal for privacy, compliance & edge use cases.

Covers partitioning, querying, trend analysis, anomaly detection & more.

Read it: blog.bacalhau.org/p/distribute...

Diagram showing Bacalhau v1.7 using AWS S3 partitioning to distribute data across multiple compute nodes. An S3 bucket is depicted sending data segments to individual servers through automated partitioning.

🎯 S3 Partitioning just got smarter!

Wrangling big datasets from S3? Bacalhau 1.7 auto-partitions + retries jobs with zero hassle.

→ Object, date, regex, substring? ✅
→ Shared + partitioned inputs? ✅

Process 1000s of files—no custom logic.

🔗 blog.bacalhau.org/p/bacalhau-v...

Graphic for Bacalhau v1.7 with a fingerprint, gear, and shield icons, illustrating the theme “Simplifying Bacalhau’s Authentication Model” on a blue background.

Bacalhau 1.7 is here and the auth game just leveled up 💥

→ Basic Auth (bcrypt optional)
→ API Tokens
→ SSO via OAuth 2.0 (Okta, Google, Azure)

Full walkthrough, sample configs, curl examples, and why this matters 👇

🔗 blog.bacalhau.org/p/bacalhau-v...

Bacalhau v1.7.0 - Day 2: Scaling Your Compute Jobs with Bacalhau Partitioned Jobs This post is part of the 5-days of Bacalhau 1.7 series.

New blog drop: Partitioned Jobs in Bacalhau v1.7 🎉

Split your job across nodes. Retry failures. Speed things up.

Big data doesn’t have to mean big pain.

Check it out 👉 blog.bacalhau.org/p/bacalhau-v...
#distributedcomputing #bacalhau

Announcing Bacalhau 1.7: Empowering Enterprises with Enhanced Scalability, Job Management, and Support (5:35) Bacalhau v1.7.0 makes distributed computing easier with new licensing, partitioned jobs, and simplified authentication.

Big news: Bacalhau 1.7 just dropped!

We’re talking a smoother dev experience, richer job feedback, and fixes that make a difference. If you’re building on decentralized compute, this one’s for you.

Read all about it: blog.bacalhau.org/p/announcing... #opensource #decentralizedcompute

The Modern Data Stack: A Scalable Future with Distributed Computing (4:03) Navigating the Challenges of the Modern Data Stack

The Modern Data Stack is evolving—scalability shouldn’t mean soaring costs. 🚀

Our latest #Bacalhau post explores how decentralized data processing helps teams scale efficiently & cut cloud costs.

Read more 👉 blog.bacalhau.org/p/the-modern... #DataEngineering #ModernDataStack

Announcing Bacalhau v1.6.5: Networking Updates & Usability Improvements (2:22) Bacalhau v1.6.5 adds networking opt-out options and usability fixes ahead of v1.7’s default networking shift.

🚀 Bacalhau v1.6.5 is here!

This update improves networking & usability while prepping for v1.7, where networking defaults to ON.

🔹 New RejectNetworkedJobs config
🔹 Default job count auto-sets to 1
🔹 Better Docker image checks

⚠️ Update configs now!

🔗 blog.bacalhau.org/p/announcing...

So, a belated shout-out to the folks at CRN for considering us, and a congrats to Red Hat on the win. This only makes us hungrier to go even bigger in 2025!

@bacalhau.org has been pushing the boundaries of distributed computing, and seeing it recognized on this level—even if we only just found out—is a big deal for us.