The thing finance devs want from AI isn't faster answers. It's defensible ones.
Take Apple's Q3 2025 revenue.
Ask any AI and you'll get three different numbers:
→ Fiscal Q3, ending June
→ Calendar Q3, ending September
→ Or either of those, converted to EUR
Same question. Three answers. Most APIs return one and don't tell you which.
We're a community partner of the NY @TechWeek_ Closing Yacht Party—and it's going to be epic. 🛥️
📍June 7 | 2–6PM | Pier 83, NYC
1,000+ founders and investors, DJs, Manhattan skyline.
Last year had 5,000+ RSVPs. Get your ticket before it sells out. Link in comments.👇
Macroeconomic analysis across all 27 EU member states. Cited. Structured. Auditable.
This is what agentic finance research looks like when you give it real data.
@LangChain + @youdotcom Finance Research API 👇
This macroeconomic research agent powered by Deep Agents, LangSmith, and the @youdotcom Finance Research API:
✅ Analyzes GDP data
✅ Detects anomalies
✅ Investigates structural & cyclical drivers at the sector level
✅ Produces structured, cited briefings
In NYC for @Techweek_?
We're hosting an AI Happy Hour on Friday, June 5, with @foundersbay, @descopeinc and @withoneai. Founders, engineers, investors, and researchers—all under one roof.
🍹 Fri 6/5 | 5PM ET | The Ivory Peacock
RSVP 👇
4 events. 1 week. NYC. 🗽
6/1: Female founders breakfast
6/4: AI Founder × VC
6/5: AI Happy Hour
6/7: Yacht Party
@youdotcom is at #NYTechWeek June 1–7. If you're building in AI, don't miss out.
See all events and RSVP at the link in the comments 👇
Tonight's VC & LP dinner is sold out!
We're bringing together 70+ of the top VCs and LPs in Silicon Valley for two panels that don't happen at every conference:
1️⃣ How VCs actually raise their own funds
2️⃣ How AI is changing LP-GP capital raising at scale.
If you missed it—more are coming. Comment if you'd be interested in attending in the future. 🔮
Finally got agent pilled and set up Hermes agent on Opus, https://t.co/ZIPvi75I7T for search, connected to whatsapp, told it to install Gstack on itself. Feels like I'm playing Productivity(TM): The Video Game. I'm addicted.
The bottleneck has shifted from capacity to curation.
a 200K window with sharp retrieval beats a 1M window full of clutter, almost every time.
Measure your effective context on your workload. not the one on the model card. the one on your stack.
https://t.co/Htwgxa7yK4
The fix isn't bigger windows. it's better context engineering.
→ retrieve the right chunks, not all of them
→ budget tokens across system prompt, history, tool outputs, output buffer
→ order retrieved content by relevance, highest first
@youdotcom is now a verified @n8n_io node 💪
Web search, deep research, and URL extraction—wired directly into your automation workflows. A proper node, not a webhook hack. ⚙️
https://t.co/3ZHL2gm25D