Pinned Tweet
🚀Build your AI engineering skills at AI Dev NYC: knowledge graphs, agent memory, harnesses, self-improving agents, inference and security. Plus a fireside chat with Andrew Ng and Yann LeCun on world models.
📅Nov 30–Dec 1 · Chelsea Piers, NYC
ai-dev.deeplearning.ai/
GPT-6 Astra edited 3.5% of a neural network’s weights in one pass.
The AI-powered car went from 0 of 256 successful runs to 1,024 of 1,024.
No explicit retraining. No tools. No gradient steps.
Read the full story in The Batch, our weekly AI newsletter 📬: deeplearning.ai/the-batch/ll…
#DeepLearningAI #MachineLearning #LLMs
4
1,100
DeepLearning.AI retweeted
It's been a week since our short course with @DeepLearningAI and @AndrewYNg went live: Tabular Foundation Models for Data Science Agents.
I taught it alongside @FrankRHutter and Eliott Kalfon. It covers four main topics:
· How tabular foundation models like TabPFN predict on a new dataset without training
· Interpreting results
· How these models are actually built (Frank even walks you through training your own nano TabPFN from scratch, on synthetic data)
· And how to hand that capability to an AI agent so it can recognise a prediction task and call the model as a tool.
In the agent lesson, we give Claude Code the TabPFN MCP server and a skill, so the agent decides when a question is a prediction problem and delegates it rather than hand-rolling a model.
If you've gone through it already, let me know what worked and what didn't, especially the hands-on labs.
If you haven't, it's free. Enroll here: bit.ly/4sQZVAA
AI coding agents write code fast, but review is the bottleneck.
@tariqshaukat, CEO of Sonar, covers agent failures and engineering fixes.
Main stage, Monday, Nov 30.
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #AICoding
Agents now use the web as much as people. But search was built for us, not them.
@saahil9jain, CTO of @youdotcom, on rebuilding search and inference for agents, and training open models to use it. AI Dev NYC main stage, Tuesday, Dec 1.
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #AIAgents
Why do AI agents keep making the same mistakes?
Build agents that improve with every run. Free short course with @Oracle
deeplearning.ai/courses/buil…
#AIAgents #Oracle
DeepLearning.AI retweeted
I'm teaching two hands-on workshops on agent harnesses at AI Dev NYC: measure, then optimize.
@AndrewYNg and the @DeepLearningAI team host it at Chelsea Piers in New York on November 30 and December 1.
A harness is the code around the model: its prompts, its tools, how it manages context and when it decides to stop.
Part 1 on November 30: "Your benchmark score is a harness score." With identical weights, Muse-Glimmer-30B costs $1.73 per solved Terminal-Bench 2.1 task in OpenCode and $8.80 in OpenClaw.
In the room, each table runs two harnesses on a 10-task slice of Terminal-Bench 2.1, and we build a shared leaderboard of pass rate and cost.
Part 2 on December 1: "Don't train the model, evolve the harness." A meta-harness is an agent that reads failed runs and patches the harness around a fixed model.
This loop took GPT-OSS-20B from 4.8% to 14.8% on held-out Terminal-Bench 2.1 for $50 of API credits. In the workshop, you run your own loop and watch the patches land.
Bring a laptop!
Thanks @cgeorgiaw for the intro and @amandamork for the invite!
Sessions: ai-dev.deeplearning.ai/speak…
Training a video generation model is a big feat, but serving it live to millions in real time brings a whole different tier of engineering challenges.
Kamil Sindi, CTO of @runwayml covers how video foundation models are trained and how real time inference scales. AI Dev NYC main stage, Tuesday, Dec 1.
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #MachineLearning
Anthropic’s S-1, by the numbers:
$16.3B revenue in H1 2026, up from $1.2B in H1 2025.
$974M operating profit in Q2.
Compute costs fell from 251% of revenue to 48%.
Claude wrote 90%+ of codebase contributions in September.
Read the full story in The Batch, our weekly AI newsletter 📬: deeplearning.ai/the-batch/an…
#DeepLearningAI #Anthropic #AIIndustry
Most agents forget what went wrong last time.
@DylanCouzon (@qdrant_engine) gives a simulated fly brain a memory of bad experiences, then tests whether it avoids new hazards. A hands-on workshop on building agent memory with retrieval at AI Dev NYC.
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #AgentMemory
Most agents are built for chat. Production asks for more – agents that work on live websites, and agents people can talk to.
Two AI Dev NYC workshops:
Idan Raman (@AnchorBrowser Browser): web agents that replay proven workflows as code
@_ashwyn Sharma (@VocalBridge): turning a chatbot into a real time voice agent
📅November 30–December 1, 2026 · Chelsea Piers, New York
🎟Get your tickets: luma.com/AIDevNYC2026?utm_so…
#AIDev #DeepLearningAI #AIAgents
LLMs reason well over text. Predicting from a table is a different job.
@tuanacelik of @prior_labs leads a hands-on workshop on tabular foundation models like TabPFN: grounded predictions on unseen tables, and where they fit inside an agent.
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #AIAgents
Basic similarity search finds text that looks relevant. It doesn't check whether that evidence applies to the context.
@JMHReif Jennifer Reif (@neo4j) & Nyah Macklin lead a hands-on AI Dev NYC workshop on closing this gap:
Diagnosing an agent citing evidence that doesn't apply
Fixing it with the simplest approach that works
Modeling connections a harder case needs
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #KnowledgeGraphs
What would you build if people could talk to your app?
Build a voice-interactive game with Ashwyn Sharma (@_ashwyn in Voice for AI Agents and Applications, our course with @VocalBridge
Enroll: deeplearning.ai/courses/voic…
Retrieval finds evidence that matches the question. It doesn't check whether that evidence applies to the context.
Nyah Macklin (@neo4j) & Jennifer Reif lead a hands-on AI Dev NYC workshop on closing this gap:
Telling relevant evidence apart from applicable evidence
Using connections across data (customer, product, date) to narrow what counts
Adding checks for when an agent has enough evidence or should stop
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #AIEvaluation
Make predictions on tabular data without training a new model for each dataset.
In Tabular Foundation Models for Data Science Agents, built with @prior_labs, you'll give an AI agent a tabular foundation model as a prediction tool.
Enroll for free: bit.ly/4sMtNOz
Ever had your retriever pull the right documents, but the AI answer was still wrong?
@steveonjava Chin (@neo4j) shows when GraphRAG beats vector only retrieval, and how to run it in production with evaluation, provenance and access control. Main stage at AI Dev NYC.
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟Get your tickets: ai-dev.deeplearning.ai/speak…
#AIDev #DeepLearningAI #GraphRAG
AI has helped create 2 million jobs, according to LinkedIn data. AI engineer is one of the fastest growing roles in 7 major economies.
This week, Andrew Ng explains why embedding AI-powered software engineers in marketing, HR, and finance now makes sense.
Read Andrew’s full letter in The Batch, our weekly AI newsletter:
📬 deeplearning.ai/the-batch/ai…
#DeepLearningAI #AIEngineering #CareerGrowth
A model is only as accurate as the context you give it.
Stephen Chin, Nyah Macklin, and Jennifer Reif of @neo4j teach grounding models with data at AI Dev NYC: a GraphRAG talk on the main stage, plus a hands-on workshop on catching when your agent trusts evidence it shouldn’t..
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟Get your tickets: luma.com/AIDevNYC2026?utm_so…
#AIDev #DeepLearningAI #GraphRAG
Gemini 4 Carbon vs. Claude Opus 5.5:
Same price per token (not counting Carbon’s intro discount).
Carbon uses 29% fewer output tokens.
Carbon has 23% lower cost per task.
Carbon is #1 on AA’s independent Intelligence Index (59).
Read the full story in The Batch, our weekly AI newsletter: 📬 deeplearning.ai/the-batch/ge…
#DeepLearningAI #AIEngineering #LLMs
Most agents forget everything between sessions.
Hear @richmondalake (@Oracle) Veljko Kovac (@cognee_) & Dennis Pilarinos (Unblocked) talk Memory & Context at AI Dev NYC!
📅Nov 30–Dec 1, 2026 · Chelsea Piers, NYC
🎟luma.com/AIDevNYC2026?utm_so…
#AIDev #DeepLearningAI #AgentMemory











