
End-to-end analytics platform with industry-leading price-performance for both cloud and on-premises workloads, built on OS @trinodb
Boston, MA
Joined October 2017
- Tweets3.9K
- Following931
- Followers3.4K
- Likes1.6K
Pinned Tweet
Meet Starburst Torin: a GPU-native analytics engine built on NVIDIA cuDF.
Run existing SQL on NVIDIA GPUs with up to 7x faster query performance in Starburst benchmarks. No changes to queries, apps, or agents.
okt.to/TloC6y
#NVIDIAGTC
Excited to be at #NVIDIAGTC Berlin!
Find Starburst at booth 3072, October 21–22, and catch CEO Justin Borgman on Thursday at 4 PM CEST for a panel on accelerated data processing.
As AI workloads evolve, so do the demands on enterprise data infrastructure. Let's talk about it!
Four cities. One clear message: Europe is building what’s next in AI.
Tomorrow, Jensen takes the stage at #NVIDIAGTC Berlin to continue the conversation.
See you there.
Berlin, we're here!
Starburst is headed to #NVIDIAGTC this week.
Find us at booth 3072, October 21–22, and catch CEO Justin Borgman on Thursday at 4 PM CEST for a panel on accelerated data processing for AI workloads.
Panel details for those interested 👀
nvidia.com/en-eu/gtc/session…
Enterprises have already paid for GPU capacity they're not fully using. We can now slot batch analytics jobs into that idle window overnight. The results? 3-4x faster queries, zero new hardware.
More here: okt.to/9nrYzo
Great event with @nvidia talking all things enterprise intelligence 🔥
Thanks for coordinating @Cvisionintl
Enterprise intelligence was the focus in Boston as CIOs, AI, and data leaders gathered to discuss trusted data, analytics, and Agentic AI. #CVisionIntl @StarburstData @NVIDIA
Starburst now runs SQL analytics on @Nvidia GPUs (via cuDF) — 3-4x faster than CPU, up to 7x on complex queries.
It started with a Vera CPU benchmark that beat Intel and landed in Jensen's keynote. Now banks' idle GPU capacity now powers faster analytics.
RSVP to join us at Starburst HQ during Boston AI Week for great food, drinks, and AI discussions.
We’ll have a talk on the bad & the ugly of shipping AI agents in production.
📅 Tue, Sep 29 | 6–9 PM ET
RSVP here: okt.to/d74Phm
#BostonAIWeek2026
We're sponsoring #BostonAIWeek 2026 🚀
From Sept 24–Oct 2, we're joining Boston's AI community for a week of conversation on where enterprise AI is headed.
See our events: okt.to/Y2FqZb
Join Starburst in Boston for a Boston AI Week meetup on the good, the bad & the ugly of shipping AI agents in production.
📅 Tue, Sep 29 | 6–9 PM ET
📍 Starburst HQ, Boston
RSVP: okt.to/ldFgWM
#BostonAIWeek
Data has gravity. The bigger it gets, the harder it is to move, so apps and tools end up orbiting it instead. Stop fighting that pull. Design for a permanently distributed data estate.
Read more: okt.to/Xs4dSq
#DataGravity #DataArchitecture #DataFederation
From Sept 24–Oct 2, we're joining Boston's AI community for a week of conversation on where enterprise AI is headed. From data integrity to agents built on real enterprise data.
See the full Boston AI Week schedule: okt.to/4FaMye
#BostonAIWeek2026
This blog explores the context layer, the tier between enterprise data and the people/agents using it.
Read the full article: okt.to/HxCdt0
#EnterpriseAI #ContextLayer #AgenticAI #DataProducts #DataFederation
Our latest blog explores business context, the rules, definitions, & institutional knowledge that turn technically correct queries into answers that are actually good for your business: okt.to/Heq2w6
#EnterpriseAI #BusinessContext #ContextEngineering #AgenticAI
Enterprise AI has a data access problem.
AI is only as smart as the data it can reach.
Our latest post lays out a practical way to start breaking that bottleneck.
Read the full article: okt.to/Z1wKb8
#EnterpriseAI #DataAccess #DataFederation #AgenticAI
Centralizing your data doesn't mean it works together. Two tables in the same warehouse can still disagree on what "revenue" means or which system owns the truth.
This article looks at fixing that with federation: okt.to/7X8M9R
#DataFederation #DataArchitecture
Your Snowflake data can be tidy and still leave AI agents guessing.
The missing piece is context. Our new post explores how a governed context layer + Starburst's federation give agents that context: okt.to/BsAqQm
#Snowflake #ContextEngineering #EnterpriseAI
Sept 16: Starburst's Pachu Shrinivas breaks down modern data architecture:
→ How the layers work together for analytics & AI
→ Where federation fits (and when it doesn't)
→ How to audit your stack: keep, connect, or consolidate
Register here: okt.to/mX0j51













