- Big Data News Weekly
- Posts
- 🤖 Building a Good Vertical Agent
🤖 Building a Good Vertical Agent
🦾Plus: Jeff Bezos' $41B Artificial General Engineer 🤖
In partnership with
Hey folks! Let’s get into Big Data and AI craziness…
In today's edition: Here’s roundup of practical playbooks, big capital moves, and tooling that will shape data reliability and AI operations 👇
📊The 2026 Data Observability Vendor Database
🧠Turn SKILL.md Into a Trainable Artifact
🌐OpenAI gives Codex the keys to Chrome's debugger
⚡How to design a multi-agent system that skips the LLM
⚽ AI suits up for soccer's biggest stage
💡 AI Tutorial:How to switch from Claude Cowork to Codex in minutes
🤖AI Tools and Data Tools to checkout
Earlier models needed custom tools and everything spelled out to work. Today's models can absorb a ton of context and reason over raw data without choking. Bigger context windows tempt people to throw in more data, but noise causes accuracy to drop. Increase accuracy by building context like a memory hierarchy.
Most AI agents demo well. Few ship real work.
Most AI agents can run a task. The problem is everything around it: setup, memory, context, cost, and figuring out what actually happened.
SureThing turns useful AI skills into autonomous agents with business context, persistent memory, cost-aware model selection, and a live dashboard. Paste a link, assign the work, and your agent reports back like a human teammate: what it did, what it cost, what needs your decision, and what happens next.
Built for founders, operators, and marketers who want AI to ship work, not become another tool to babysit.
The data observability market has evolved rapidly over the past five years. What began as a niche category focused primarily on monitoring modern data pipelines has expanded into a broad ecosystem encompassing anomaly detection, data quality, lineage, schema monitoring, business observability, and increasingly, AI-driven analytics
Fine-tuning an LLM is often expensive and slow. SkillOpt offers a different solution: instead of training the model, it trains the document the model reads. SkillOpt is a text-space optimizer for agent skills that treats your SKILL.md file as the trainable parameter and improves it through a feedback loop. No GPU. No weight updates. The model never changes.
The ChatGPT maker just rolled out Developer Mode for browser use in Chrome and the Codex in-app browser. It gives the coding agent access to the Chrome DevTools Protocol. This lets it profile JavaScript performance and inspect console output, network traffic, and page state to track down bugs.
You’ll learn how to build multi-agent systems that are both highly scalable and cost-effective. You’ll also get a handle on designing hybrid architectures that balance AI with deterministic code. By using callbacks, you can bypass unnecessary LLM requests and implement Redis session storage to run thousands of agents at the same time.
10x the context. Half the time.
Speak your prompts into ChatGPT or Claude and get detailed, paste-ready input that actually gives you useful output. Wispr Flow captures what you'd cut when typing. Free on Mac, Windows, and iPhone.
👨💻 Data Tools, Libraries
xloil
xlOil provides framework for interacting with Excel in different programming languages (python & C++ currently)
Cakework
Build serverless async backends without cloud resources.
DriftDB
A real-time data backend for browser-based applications.
Sintra: The AI "dream team" for overwhelmed operators. Access 12+ specialized agents that handle marketing, sales, ops, and much more for you - no prompting experience required.
AI News:
Amazon founder Jeff Bezos just shed more light on his AI startup Prometheus while announcing a $12B round at a $41B valuation, pitching an "artificial general engineer" for physical machines — and also waving off AI job-loss fears. Bezos started the company in late 2024 alongside Vik Bajaj, a physicist and chemist who helped create Verily, Alphabet's life-sciences arm.
From idea to shipped tool in 11 minutes.
Type the problem in Slack. Viktor writes the code, deploys to your subdomain, posts the URL, and starts using it on your next request. No specs, no Jira, no kickoff. Founders are running entire companies this way.
The 2026 FIFA World Cup opened in Mexico City with AI wired into nearly every layer of the tournament, as companies like Lenovo and Google integrate the tech into areas like offside calls, team analytics, fan experiences, and more. An optical tracking system will capture more than 150M data points per match, along with a motion-tracking Adidas ball that reports 500 times a second.
SpaceX priced its IPO at $135 a share, near $1.75 trillion valuation, and opens today on Nasdaq as SPCX. Reuters reported demand crossed $250 billion. But look past the rocket headlines and you're mostly buying AI. SpaceX took over xAI in February, so it now runs Grok, sells $1.25 billion a month in compute to Anthropic, and just revealed AI1, a solar satellite designed to run a data center in orbit.
Microsoft has published a deep look at how its Copilot Studio team validates AI evaluation systems, addressing a growing enterprise concern: if you're using AI to grade your AI agents, how do you know the grader is right? The answer, according to Microsoft's data science team, involves treating evaluators like any other model that needs testing.
The agentic-finance race is heating up. Coinbase debuted AI agents that can execute trades and pay for premium research on a user's behalf just days after Robinhood rolled out trading agents of its own. Users can connect the agent to their main account or run it in a walled-off sandbox, and ask it to rebalance a portfolio or trade against a thesis; it works inside ChatGPT or Claude via Coinbase's MCP server.
How do companies like Netflix, Airbnb, and Doordash apply machine learning to improve their products and processes? We put together a database of 200 case studies from 64 companies that share practical ML use cases and learnings from designing ML systems.️
AI Tutorial
How to switch from Claude Cowork to Codex in minutes
You don’t even need to pay for the ChatGPT agent. You can use Microsoft Copilot for free. Here is how:
Go to the Codex and download the Codex for your computer.
Open Codex and sign in with your ChatGPT account.
During setup, Codex may ask if you want to import your setup from another agent.
Choose Claude Cowork if it appears on the screen.
Select the items you want to move over, such as tools, setup, projects, and recent chat sessions.
Click continue and let Codex bring everything across.
Once it finishes, your Codex workspace should already have your old setup ready to use.
🔥Top AI tools to increase productivity:
Thor.ai captures the blockers, decisions, and commitments that never make it into your tools — straight from Slack threads and meetings
AirRankPilot helps local businesses get discovered by Google and AI tools like ChatGPT and Perplexity
InboxKit is your all-in-one platform for building, managing, and scaling cold-email infrastructure.
CCPayment is a cryptocurrency payment platform allowing merchants to accept and payouts
HuePress is a SaaS platform providing therapy-grade, high-quality printable coloring pages
Creatives Takeover is an AI platform that turns raw founder ideas into business plans
Pixel‑exact AI image generator composing scenes on fixed canvases
View our database of all the best AI tools for your needs: aitoolsup.com
Have cool resources to share? Submit AI tool
A.I. Generated Image of the Day
👀 Ancient God
Recommended reading:
SPONSOR US
Get your product in front of Big Data & AI enthusiasts
😮 The Marketing Channel You've Probably Slept On:
Haven't tried newsletter sponsorships yet? You are missing out on a HUGE ROI+ customer acquisition channel. I know because dozens of advertisers keep coming back for more... Run a test campaign with us and see for yourself 👉 Get in touch today.
What did you think of today's email?Your feedback helps me create better emails for you! |