dennylee

geek, scribe, coffee snob, and wanna-be cyclist. Contributor to Apache Spark and Delta Lake maintainer. Developer Relations at Databricks (opinions r my own)

[Daniel Lenton @DanielLenton1]

OpenClaw, Hermes Agent and Prime Agent don't meaningfully learn from experience.

On Continual-ARC, our new benchmark at @letsunifyai, none of them were able to memorize anything of value and none were able to generalize to new tasks. Training a small 9B model's weights did both.

[Databricks @databricks]

🌁 Heading to AI Engineer Code Summit? Come build with us!

Tomorrow, our own Elise Gonzales and @bradfordc are hosting a hands-on @omnigent_ai workshop.

You'll build a GitHub CI workflow that connects specialist agents across models and coding harnesses to review PRs, coordinate repairs, and validate changes. Leave with version-controlled agent configurations and a runnable workflow you can adapt to your own projects.

Workshop details ↓
https://t.co/1IgiKX9PAQ

[Ankit Mathur @ankit_math]

I’m speaking at the @aiDotEngineer about agent infra for the next generation of agents!

Join me Wednesday, 11:45AM on the main stage

Session details here: https://t.co/tJVlGN9A7j

[Ali Ghodsi @alighodsi]

Day 0 launch of Gemini 4 Carbon on @databricks! You can just get the tokens from Unity Gateway on Databricks.

[Ali Ghodsi @alighodsi]

Day 0 launch of Claude Fable 5.2 on @databricks! You can just get the tokens from Unity Gateway on Databricks.

[Omnigent @omnigent_ai]

Come build with us at AI Engineer Code Summit!

Elise Gonzales and @bradfordc are hosting a hands-on Omnigent workshop. 🛠️ You'll build a GitHub CI workflow where specialist agents review PRs and validate changes.

🗓️ Tues, Nov 10 📍 San Francisco

Workshop details👇
https://t.co/45D0wTZCFe

[Databricks @databricks]

Claude Fable 5.2 lands on Databricks on Day 0.

Built for coding, finance and science, it sets a new high on our OfficeQA Pro v2 benchmark.

Run long-running agents on governed data, with Unity Gateway controlling calls across 70+ models.

[Databricks @databricks]

We started by giving engineers access to the most expensive AI coding models and as many tokens as they wanted. Adoption took off. So did costs.

At Databricks, we've since reduced AI coding unit costs by up to 90% in some cases through more efficient models, better routing, spend visibility and less token overhead.

Our co-founder @pwendell explains what worked and how we kept developers moving. https://t.co/JPWw2pn3UL

[Daniel Liden @danjliden]

In case you missed it, @omnigent_ai 0.21 was released last week with:
- easy-to-read tool calls and results
- compact PR lists and inline reviews in the PR viewer
- SSH sessions from the desktop app
- model/reasoning effort controls for side chats
- much more

[Omnigent @omnigent_ai]

At QCon San Francisco, @ev_hur and @dennylee will present Multi-Agent Coding with Omnigent.

In this talk, they will cover how Omnigent works under the hood, how they use it to automate their own workflows, and what building it in the open has taught them.

🗓️ Wednesday, November 18
👉 Session details: https://t.co/66R0abj6ew

#Omnigent #OpenSource

[Yuchen Jin @Yuchenj_UW]

DeepSeek-V4.1-Flash-1017 is an amazing model.

DeepSeek is one of the few OSS model labs that treat inference efficiency as a first-class principle in model architecture design, then open-source all their “secrets.”

S-tier lab.

[Databricks @databricks]

A great Genie demo is just the beginning. Here's how to turn early momentum into trusted, organization-wide adoption.

→ Start with one team and a clear set of business questions
→ Test answers against known-correct results
→ Refine business definitions based on feedback
→ Expand to adjacent teams as accuracy and governance checks are met

Read the step-by-step guide: https://t.co/HSJKtCNS72

[Brett Taylor @Brett_A_Taylor]

You think you're about to read some idiot who believes he could come off the couch and compete against the best in the world.

Instead, you read something brilliant:

[Zohran Kwame Mamdani @ZohranKMamdani]

It's morning again in America.

[Peter Daou @peterdaou]

I've been in politics for a quarter century. It's hard to express how staggering it is to see a conversation like this on corporate US media. The Morning Joe panel gobsmacked by the directness of @AbdulElSayed's points.

(clip via @DropSiteNews)

[Databricks @databricks]

Data prep doesn't have to start with code.

Lakeflow Designer lets you prepare and transform data directly in Databricks using a visual, no-code interface.

Great walkthrough from @joshbogran.

[Databricks @databricks]

Agents can build software. But what about the databases behind it?

Databricks VP of Engineering and @neondatabase co-founder @nikitabase joined @MTSlive to talk about database branching, serverless architecture, and building for both AI agents and enterprise workloads.
https://t.co/jlr2hmFmi1

[Michael Grinich @grinich]

An update on the WorkOS security incident:

Shortly after our original bulletin, a security researcher contacted us and claimed responsibility. They told us they were evaluating ARTEX, an autonomous penetration testing framework, when their agent escaped its isolated testing environment and interacted with WorkOS unintentionally.

We asked for technical evidence. The researcher provided request details, timestamps, and artifacts from the activity. Our security team and an independent forensic investigator matched that evidence against our logs, establishing with high confidence that this individual was responsible for the unauthorized activity we observed.

The researcher has confirmed that all WorkOS data obtained by their agent has been deleted and was not shared with anyone else.

Their explanation that the activity was unintended does not change how seriously we take the incident or our responsibility to our customers.

Full report: https://t.co/gfKc1HCQ81

[Databricks @databricks]

Introducing Apache Spark® 4.3, now available in Databricks Runtime 19!

The release brings major new capabilities across real-time processing, history-preserving pipelines, small-data performance, Spark Connect, and SQL.

What’s new:

- Bring state to Real-Time Mode with streaming aggregation and transformWithState
- Preserve history with SCD Type 2 Auto CDC
- Iterate faster on small data with Project Feather
- Reach Spark from more applications with expanded Spark Connect capabilities

It also begins a new quarterly feature-release cadence, bringing completed improvements to users sooner.

Get started: https://t.co/FF22P4x06r

[Ali Ghodsi @alighodsi]

This is an awesome video on how Lakeflow Designer works! If you use Alteryx, then this is a must see video!