Sign in

Tigris Data

@tigrisdata.com
228 followers 375 following 443 posts

Tigris is a globally distributed S3-compatible object storage service that provides low latency anywhere in the world. | tigrisdata.com | Based in SFO

PostsRepliesMedia
Tigris Data @tigrisdata.com · 22/09/2026
If a job finishes instantly, that's great. The job reaches message queue enlightenment and re-enters the cosmic background radiation until its time is needed next. If not, the job is trapped in a near endless cycle of rebirth bouncing between workers until one doesn't crash this time, we hope.
100
Tigris Data @tigrisdata.com · 15/09/2026
So Xe invented a new brave format that does store enough information to construct ranged get requests. This ended up making Git object storage on top of Tigris object storage upwards of 4.5-14.6 times faster than the old approach of badly adapting filesystem interfaces to Tigris.
151
Tigris Data @tigrisdata.com · 15/09/2026
This is partially a lie because Git tracks the _entire file_ at any given point in history. Every copy of all of those versions of all of those files are stored in objects which are named after the checksum of their contents (content-aware storage).
120
Tigris Data @tigrisdata.com · 21/08/2026
An agent can run off a VM image that is larger than the disk it runs on. AgentENV by KVCache.AI keeps microVM disk images in object storage and pulls down only the blocks each machine actually needs. That is how image size stops being limited by local disk.
110
Tigris Data @tigrisdata.com · 18/08/2026
How does Tigris work? What are the moving parts? Today we have a transcript of our CTO Himank's talk about how Tigris works at a high level and how FoundationDB makes it possible.
100
Tigris Data @tigrisdata.com · 11/08/2026
Delete doesn't have to mean the data is gone. Tigris lets you un-delete your most precious objects and buckets if you(r agent) didn't really mean to actually remove it for others.
100
Tigris Data @tigrisdata.com · 30/07/2026
If you use RAGFlow by InfiniFlow to answer questions from your documents, everything it knows lives in its storage backend. By default, if you use MiniIO, RAGFlow re-reads your corpus every time you tune it.
100
Tigris Data @tigrisdata.com · 29/07/2026
@apache.org OpenDAL fully supports Tigris Data. This storage layer with one API across 50+ backends, in 17 languages. Tigris config is just three lines. Get the benefit of multi-region workers, with no cross-region egress charges and no per-region bucket topology to maintain.
100
Tigris Data @tigrisdata.com · 28/07/2026
1/ The era of manual install scripts for Blessed Frameworks™ is over. New post on `tigris init --agent` and why agent-native onboarding is becoming the baseline.
110
Tigris Data @tigrisdata.com · 23/07/2026
If you use dlt by dltHub to move data, you can set up the pipeline to save your results into a cloud storage bucket that everything else reads from. Most teams put that bucket on AWS, right next to the pipeline. However, when something reads the data back, AWS charges an egress fee.
100
Tigris Data @tigrisdata.com · 22/07/2026
If you self-host ClickHouse, you need to avoid paying egress every time your cluster reads its own data back. The standard production playbook assumes an S3 bucket sitting next to your cluster. So you point at AWS, and every restore, audit query, and backfill turns into a bandwidth bill.
100
Tigris Data @tigrisdata.com · 21/07/2026
We benchmarked @TigrisData vs S3 vs R2 on 10M small objects...agent state, checkpoints, logs etc. p90 reads: 7.9ms (Tigris) vs 42ms (S3) vs 681ms (R2)Sub-10ms means object storage stops being an archive and starts being your app's working memory. Full report: www.tigrisdata.com/blog/benchma...
010
Tigris Data @tigrisdata.com · 10/07/2026
Object storage finally branches like Git We helped ComputeSDK build storagesdk. With this universal SDK, you can fork a bucket, let an agent or experiment run wild, then keep it or throw it away. Your production data never moves and the agent knowledge base is secure.
110
Tigris Data @tigrisdata.com · 07/07/2026
Strip an agent to parts: model, harness, sandbox, state. Three are replaceable. The state IS the agent. And today it's scattered across Postgres, Pinecone, Redis, LangGraph, and Temporal. You can't checkpoint an agent that lives in five places.
100
Tigris Data @tigrisdata.com · 03/07/2026
Object storage finally has an undo button: tigrisdata.com/blog/soft-de... You can turn on soft delete and every delete drops into a holding state for 7–90 days.
100
Tigris Data @tigrisdata.com · 02/07/2026
Tigris is happy to be partnering with Isorun to bring persistent state to their agentic runtime. Isorun gives agents fast, hardware-isolated compute. We give them durable, forkable disks, so sandboxes stay disposable, and the state that matters survives.
110
Tigris Data @tigrisdata.com · 01/07/2026
We built a self-updating knowledge base with Tigris! It reads your feeds nightly and posts a digest to Slack. Our product gives agents durable storage they can write to unattended, where a failed run can never corrupt what's already there.
100
Tigris Data @tigrisdata.com · 24/06/2026
Hooking an agent up to your object storage usually means handing it credentials and hoping it behaves. That's a lot of trust for something that occasionally invents a delete. storagesdk ships an MCP server, so storage shows up as tools an agent can call.
100
Tigris Data @tigrisdata.com · 18/06/2026
Fifty agents, one bucket. They can't share a bucket without overwriting each other, and copying it fifty times is slow and expensive. StorageSDK gives object storage git semantics. Snapshot a bucket, then fork it. A fork is copy-on-write: constant time, shares the parent's bytes.
100
Tigris Data @tigrisdata.com · 17/06/2026
Every object store has its own CLI. AWS, wrangler, gcloud, az. Tigris and ComputeSDK built one CLI for all of them: github.com/storagesdk/s... List a bucket, switch providers by changing one word. It pipes like a real Unix tool.
110
Tigris Data @tigrisdata.com · 15/06/2026
Agents sometimes delete the wrong thing. That's no longer a problem: www.tigrisdata.com/blog/soft-de... We just added Soft Delete to Tigris. You can flip it on once, pick a retention window, and every delete just becomes recoverable.
110
Tigris Data @tigrisdata.com · 11/06/2026
S3, R2, GCS, Azure all great storage options - but can they fork buckets? Turns out storage for agents is different, and our compute-based friends need to experiment and drop buckets at will. storagesdk fixes that with one idea: snapshots and forks as first-class operations.
200
Tigris Data @tigrisdata.com · 10/06/2026
Agents can easily manage S3-compatible storage for their own experiments via agent-kit. When you kick off an agent, @tigrisdata.com hands it its own private bucket and a scoped key just for that run, so it has a clean place to work without you wiring up any storage.
100
Tigris Data @tigrisdata.com · 04/06/2026
We've also included a brand new interface to object storage: simplestorage. This acts much closer to our TypeScript SDK while still giving you all the ergonomics that your apps/agents crave.
110
Tigris Data @tigrisdata.com · 04/06/2026
Attention Go programmers: we have something for you for once! We've brought our buckety goodness over to Go with our storage-go library! No more having to attach headers in order to rename/move an object, just call RenameObject!
141
Tigris Data @tigrisdata.com · 04/06/2026
Object storage finally branches like Git: www.tigrisdata.com/blog/storage... We built storagesdk with @computesdk. With this universal SDK, you can fork a bucket, let an agent or experiment run wild, then keep it or throw it away. Your production data never moves: github.com/storagesdk/s...
010
Tigris Data @tigrisdata.com · 03/06/2026
A fork is a metadata operation and not a data migration. You can fork your data for an agent run, let it mutate files safely, replay later from the same baseline. Git branching, but for storage:
000
Tigris Data @tigrisdata.com · 27/05/2026
When agents use object storage, they should work off of forked buckets, not the original production bucket. This way, any bad write can be quarantined.
101
Tigris Data @tigrisdata.com · 26/05/2026
We just shipped multiple lifecycle rules per bucket for Tigris: www.tigrisdata.com/blog/lifecyc... You can route heavy ML datasets to Glacier while auto-deleting stale logs on autopilot. No more global policies or janky deletion scripts.
110
Tigris Data @tigrisdata.com · 21/05/2026
We embedded a Tigris client on our homepage so that if you're in a pinch and need to fork a bucket, we've got your back!
110
Tigris Data @tigrisdata.com · 20/05/2026
@basicmemory.bsky.social just shipped per-file diff/merge for AI context as markdown. Pick any prior version, see changes, pick the lines you want. Built on Tigris bucket versions: every edit = a new object version, with full revision history from one ListObjectVersions call. No Git required 👇
110
Tigris Data @tigrisdata.com · 19/05/2026
Most AI memory locks your context to one platform. Switch tools, lose everything. @basicmemory.bsky.social flips it: your knowledge graph as plain Markdown in your own bucket editable directly. Claude, ChatGPT, etc reads and writes the same files. Read how a four-person team built it on Tigris. 👇
211
Tigris Data @tigrisdata.com · 18/05/2026
We caught the disco bug 🪩✨
000
Tigris Data @tigrisdata.com · 13/05/2026
1. Main bucket stays clean 2. New docs land in a fork 3. Sketchy → quarantine 4. Clean → promote He calls it "The Immutable Agent" 👇 www.tigrisdata.com/blog/immutab...
000
Tigris Data @tigrisdata.com · 13/05/2026
Prompt injection via the knowledge base is the agent attack surface nobody's talking about. David Myriel built a TypeScript agent on Mastra.ai using Tigris bucket forking as a last line of defense:
110
Tigris Data @tigrisdata.com · 12/05/2026
The Immutable Agent is a pattern for protecting agents that write back to shared storage: t.ly/fIbfl 1. Every agent run executes against a fork of your storage bucket, not the live one. 2. A validator checks the fork. If it passes, it merges into production. If it fails, it gets quarantined.
100
Tigris Data @tigrisdata.com · 07/05/2026
Wake up to a fresh LLM Wiki of everything from yesterday. One Slack link, and satisfy your FOMO. We built this on Karpathy's pattern. Runs nightly, costs $0.23/run: www.tigrisdata.com/blog/self-up...
000
Tigris Data @tigrisdata.com · 29/04/2026
Coordinate millions of agents without polling. Agent Kit lets you set up webhooks with a single function call—triggering agents on object writes. Event-driven workflows, built directly on storage.
020
Tigris Data @tigrisdata.com · 22/04/2026
Tigris now gives you point-in-time recovery — snapshot your agent's state, and if something goes wrong, fork back to the last good checkpoint instantly. Zero-copy, no downtime, original state preserved for debugging. Read the docs: www.tigrisdata.com/docs/use-cas...
000
Tigris Data @tigrisdata.com · 14/04/2026
Where LanceDB handles vector search, DuckDB handles SQL. Also, when an agent misbehaves, you can pin a snapshot and replay exactly what happened last at no cost.
000
Tigris Data @tigrisdata.com · 24/03/2026
Tigris just shipped Multi-Region Buckets. You can pick exactly where your data lives and how it replicates: Multi-region for resilient production workloads across the US or EU, Dual-region when you need to pin to exact region pairs, and Single-region for cost-efficient local workloads.
100
Tigris Data @tigrisdata.com · 25/02/2026
Agents re-read datasets— a lot. The "data leaving storage" is real. @crewai.bsky.social + Tigris fixes that. • Unlimited reads, no cost penalty • Durable workspace across agents + runs • Snapshots for reproducible state Storage built for how agents actually behave. Guide 👇
100
Tigris Data @tigrisdata.com · 24/02/2026
AI agents break most infra assumptions. They pause. Resume. Fork. Snapshot. Move regions. That’s not stateless microservices. We’ve been working with @agentuity.com as they build a new cloud for AI agents — and their storage model looks very different
111
Tigris Data @tigrisdata.com · 12/02/2026
Building a hosted MCP server? Auth shouldn't be the hard part. We open-sourced our MCP OIDC provider, a minimal, vendor-neutral OIDC layer that works with real MCP clients (Claude, Cursor, etc.) and any IdP. Blog + code 👇
110
Tigris Data @tigrisdata.com · 10/02/2026
If you're a developer, you live in the command line. Tigris' new CLI lets you supercharge your shell with bucket forking, object management, and more all from one easy to use program. It's just an npm install away: npm install -g @tigrisdata/cli
121
Tigris Data @tigrisdata.com · 20/01/2026
Fun fact: object storage (AKA: cloud based unlimited FTP server) is actually a key-value store under the hood! There's nothing stopping you from using it as one! Today's blogpost covers the gorey details about how and why you'd want to do that.
100
Tigris Data @tigrisdata.com · 14/01/2026
More teams are moving out of Big Cloud to neoclouds better built for AI workloads. How does the growth of neoclouds change the storage layer? Our CEO talked with @thenewstack.io about the new cloud and how distributed computing is taking over and what that means for storage. 🔗👇
111
Tigris Data @tigrisdata.com · 08/01/2026
What a year! From $25M funding to Git-like bucket forks, our 2025 was all about powering modern AI workflows with fast, flexible object storage. Catch the highlights! 👇
182
Tigris Data @tigrisdata.com · 06/01/2026
Chunking is a primary control knob for RAG quality. Compare chunking strategies using a reproducible pipeline: • @qdrant.bsky.social for per-experiment indexing and retrieval diffs • Tigris (Forked Buckets) for dataset versioning • CrewAI to orchestrate ingest → chunk → embed → evaluate Code 👇
111
Tigris Data @tigrisdata.com · 01/01/2026
Tigris lets you fork your buckets like you fork your code. How did we do it? We made our data plane append-only. To do this you have to make your log (of objects inserted) into the database.
110