Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
pierrebrunelle
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
Infinite-memory, Context-aware, & Multimodal Chatbot
(github.com)
3 points
by
pierrebrunelle
1y ago
|
1 comments
2.
▲
by
pierrebrunelle
1y ago
Building an infinite-memory, context-aware, multimodal chatbot is as easy as 600 lines of Pixeltable code. Why? Because Pixeltable is the only multimodal data infrastructure that unifies storage and orchestration for your AI workloads.
3.
▲
by
pierrebrunelle
1y ago
Agree...as simple as: @pxt.query def search_documents(query_text: str, user_id: str): sim = chunks.text.similarity(query_text) return ( chunks.where( (chunks.user_id == user_id) # Metadata filtering &
4.
▲
by
pierrebrunelle
1y ago
I fully agree and this is why context engineering matters and unifying storage and orchestration and treating agents as just another function call is significant and getting full visibility into the pipeline to easily iterate and with with
5.
▲
by
pierrebrunelle
1y ago
It really depends on the use case. For Pixelagent, I don't really look at competition as mentioned by the fact that this is a reference implementation. People should build their own; it's easy. What's difficult is the underly
6.
▲
by
pierrebrunelle
1y ago
It all started with https://github.com/pixeltable/pixeltable which is the core project we are actively working on and the problem space of working with video and frames for Computer Vision teams who struggle with the e
7.
▲
by
pierrebrunelle
1y ago
I like PocketFlow. You beat me on the # of lines of code! But does it provide parallelization, caching, orchestration, versioning, observability, lineage, multi-modal support? As you just showed, building an agent SDK is easy, so what'
8.
▲
by
pierrebrunelle
1y ago
Congrats! Take whatever you want :) Yes: Memory, observability, versioning, and lineage being built-in is a derivative of unifying orchestration and storage: https://docs.pixeltable.com/docs/datastore/computed-colu
9.
▲
by
pierrebrunelle
1y ago
Thanks! https://github.com/pixeltable/pixelagent/tree/main/examples/... for multiple tools/agents. A tool is just a UDF. You can have as many as you want. The goal of this reference agent SDK i
10.
▲
by
pierrebrunelle
1y ago
You can indeed turn anything that you want into an MCP server, e.g. https://github.com/pixeltable/pixeltable-mcp-server . Pixelagent is a reference implementation for a multimodal agent framework to show that an agent c
11.
▲
Show HN: Pixelagent – Build your Stateful Agent Framework in 200 lines of code
(github.com)
30 points
by
pierrebrunelle
1y ago
|
14 comments
12.
▲
Pixeltable-yolox – YOLOX that works with modern Python
(github.com)
1 points
by
pierrebrunelle
1y ago
|
1 comments
13.
▲
by
pierrebrunelle
1y ago
We got sick of fighting with YOLOX's dependency every time we needed to update the integration for a decent object detector demo, so we forked it and fixed the annoying issues. pip install pixeltable-yolox Key stuff fixed: - Works with
14.
▲
by
pierrebrunelle
3y ago
ChatGPT directly into Jupyter Notebook with Notebookgpt by Noteable, notebookgpt.com as a plugin: http://notebookgpt.com/
15.
▲
by
pierrebrunelle
3y ago
Also a different one: https://github.com/noteable-io/genai with additional magics.
16.
▲
by
pierrebrunelle
4y ago
Nothing breaks the flow of experimentation in a notebook like an exception. You’re deep in thought, pulling data, and then suddenly everything comes to a halt. To empower you to fix exceptions quickly we're releasing genai: an open-sou
17.
▲
Bringing GPT to Interactive Python
(github.com)
6 points
by
pierrebrunelle
4y ago
|
1 comments