Pixeltable
@pixeltablehq
Followers
138
Following
105
Media
20
Statuses
111
AI Data Infrastructure — Declarative, Multimodal, and Incremental https://t.co/YzWK3wqlz4
San Francisco
Joined August 2024
Pixeltable: AI Data Infrastructure Providing a Declarative, Incremental Approach for Multimodal Apps.
0
4
109
RT @pjlbrunelle: @3scorciav @pixeltablehq @Voxel51 Great that you are bringing it up, we actually have an integration with Voxel51: https:/….
docs.pixeltable.com
Build image analysis workflows with Pixeltable and Voxel51 in two phases
0
1
0
RT @pjlbrunelle: @rown If you're still in the midst of architectural decisions for your multimodal data stack, have a look at @pixeltablehq….
0
1
0
RT @pjlbrunelle: That's all you need to look at to know who's winning. Delivered to you by @OpenRouterAI
0
3
0
RT @pjlbrunelle: @pixeltablehq is not merely another open source tool or library to be added to an already complex stack; it is foundationa….
0
1
0
RT @pjlbrunelle: Go from AI experiment (ML/CV/LLM) to production pipeline in 1 second. >Experiment:.table. select(openai=openai_call, mist….
0
1
0
RT @pjlbrunelle: How to build multimodal production-grade MCP Servers. Walkthrough by @moteropedrido. Link below 👇
0
3
0
RT @pjlbrunelle: I keep seeing teams wrestling with the same data annotation problem: Label Studio, Scale Nucleus, Labelbox, CVAT are great….
0
1
0
RT @arazvant: Free hands-on course on building an AI Agent that can understand Videos. What's inside:.|.⇢ Define structured multimodal pip….
0
3
0
RT @phineasb: The customer is always right! (Thanks @pjlbrunelle and the whole team @pixeltablehq - feeling is mutual).
0
1
0
Get the code: Try the notebook:
github.com
Pixeltable — AI Data infrastructure providing a declarative, incremental approach for multimodal workloads. - pixeltable/pixeltable
Build a production-ready multimodal AI application that combines Object Detection (ML) and LLM (e.g. OpenAI) in 2 minutes. No spaghetti code. Just a few lines of Python. Data storage, versioning, caching, lineage, and more. Here's how 🧵.
0
0
1
RT @pjlbrunelle: In Pixeltable, all media (videos, images, audio) resides in ext. files, and PXT stores references to those. Files can be….
0
1
0
RT @pjlbrunelle: Context Engineering -> Agents should be Data-Centric Applications.Agents = LLM + Data + Orchestration. Example below ⏬ htt….
0
1
0
Create Tables. Add Computed Columns. Build Queries. As simple as that.
Context Engineering -> Agents should be Data-Centric Applications.Agents = LLM + Data + Orchestration. Example below ⏬
0
0
1
A Deep Dive into Embedding Indexes:
github.com
Pixeltable — AI Data infrastructure providing a declarative, incremental approach for multimodal workloads. - pixeltable/pixeltable
Vector search is amazing until your data changes. Then you're stuck with:.😫 Stale embeddings.💸 Expensive full rebuilds.🤯 Complex sync pipelines. See image for the fix :) -> pip install pixeltable
0
0
1
RT @pjlbrunelle: Vector search is amazing until your data changes. Then you're stuck with:.😫 Stale embeddings.💸 Expensive full rebuilds.🤯 C….
0
2
0
0
1
0
RT @pjlbrunelle: @realsohamparekh @pixeltablehq Besides Multimodal is also a pain today for AI Agents. Pixeltable makes it way easier by….
lu.ma
🚀 Build Production-Ready Multimodal AI Agents — Without the Complexity Join our 2nd Startup Tech Talk with Pierre Brunelle (CEO & Cofounder, Pixeltable) Tired…
0
1
0
RT @pjlbrunelle: @realsohamparekh Good luck! You gonna need a Data Infra that actually handles videos/images/audio then! @pixeltablehq is b….
0
5
0