This has been one of the biggest weeks in AI Agents. Here is...

@adamsilverman
Adam Silverman (Hiring!) 🖇️@adamsilverman
93 views Aug 03, 2025 ~3 min read
Advertisement
1
This has been one of the biggest weeks in AI Agents.

Here is everything that happened this week in AI Agents from AgentOps, E2B, LlamaIndex, Subconcious, Manus AI, Lovable, Potpie, Cyberdesk, Copilot kit, Camel AI, Y Combinator, & more. 🧵

(save for later)
Media image
2
1/
🚨@e2b raises $21M Series A to build cloud runtime for AI agents

We have been long time supporters of @mlejva + team. Excited to see this!


@e2b
E2B@e2b
We just raised a $21M Series A to build the cloud runtime for AI agents.

More than 88% of the Fortune 100 is already using E2B, together with companies like Hugging Face, Perplexity, Groq, LMArena or Manus.

E2B is delivering the Cloud for AI Agents.

We are hiring the A-team in San Francisco and Prague - Check out our open roles.

Thanks to the amazing support from
@insightpartners, with participation from @DecibelVC,@seedtosunflower, @kaya_vc, and angels like @scottcjohnston, former CEO of Docker, as well as our customers, partners and open-source community, we are making this happen.

Read our CEO @mlejva's take on what's next for us.

🪐 Onward.
3
2/
@AgnoAgi + @AgentOpsAI makes intelligent caching feel like regular Python

With @AgnoAgi, you just:
- Use self.session_state[key] = value to cache
- Have built-in persistence across workflows

@n_sri_laasya


@n_sri_laasya
Sri Laasya Nutheti@n_sri_laasya
Building Smart Caching is EASY

With @AgnoAgi, you just:
- Use self.session_state[key] = value to cache
- Check if `self.session_state.get(key):` for hits
- Built-in persistence across workflow runs

@AgnoAgi makes intelligent caching feel like regular Python
4
3/
ManusAI launches Wide Research

@manusai's agent spawns parallel instances for deep analysis, map-reduce & supercomputing via chat.


@ManusAI
Manus@ManusAI
Introducing Wide Research
5
4/
@elevenlabs revamps Conversational Agents page!

New direction blending simplicity, research & sound aesthetics. @RomaTesla

x.com/RomaTesla/stat…
6
5/
✨ICYMI AG-UI's Launch week

Protocol for agent-user interactions with integrations for LangGraph, CrewAI & more. New CLI, Pydantic support. @CopilotKit

x.com/CopilotKit/sta…
7
6/
High agency vibes!

@n_sri_laasya shares some agent tools she built in 4 months at @AgentOpsAI!


@n_sri_laasya
Sri Laasya Nutheti@n_sri_laasya
"I'm very high agency"
Me: I built all these in <4 months
Media image
8
7/
@SuperDesignDev hits 2k stars in 4 weeks! Cooking big updates for PM/Design/Growth folks. @jasonzhou1993


@jasonzhou1993
Jason Zhou@jasonzhou1993
Woah 😮

@SuperDesignDev just past 2.0+k stars in 4 weeks!

And the growth is not slowing down

Cooking something big for PM/Design/Growth people next

DM me if you wanna early access!
Media image
9
8/
@jerryjliu0 shares @seldo's talk on hybrid workflows, autonomy vs structure & debuggable orchestration. 🎬 @llama_index


@jerryjliu0
Jerry Liu@jerryjliu0
Effective Agent Design Patterns in Production 🤖

Building agents is a bit complicated when you peel back the underlying implementations, but this is one of the easiest resources to learn how it all works under the hood.

This is a fantastic introductory video by @seldo to the core principles and workflow architectures behind agents.
1. Build an agent to make sense of large volumes of unstructured data in a semi-autonomous way.
2. An agent is powered by an underlying workflow. Some of these patterns were originally laid out by @AnthropicAI - chaining, routing, concurrency, and more.

Video: youtube.com/watch?v=72XxWk…
Media image
10
9/
@guohao_li's introduces @Eigent_AI – open-source app handles planning, bookings & analysis in parallel. Custom workers, MCP tools & more.

x.com/guohao_li/stat…
11
10/
@SubhaghV built a WhatsApp AI agent to replace customer support teams

Sold for $2k + $500/mo. n8n + WhatsApp API build

x.com/SubhaghV/statu…
12
11/
Subconscious: Platform for long-horizon agents

One API call for tool use & extended context. Backed by MIT research. @thejackobrien


@thejackobrien
Jack O'Brien@thejackobrien
Today we're launching Subconscious: a new platform for building agents with long-horizon reasoning and tool use, backed by MIT research.

One API call. Tool use. Context beyond existing limits.

If you're building agents, let's talk.
13
12/
Doe (@DoeLabs) embeds AI agents into multi-site healthcare groups, starting with DSOs. They plug into a group's PMS, ERP, email, & supplier catalogs to complete tasks, deliver immediate EBITDA lift

Congrats @AdrianBarbir & @YuhuangOu! @ycombinator


@ycombinator
Y Combinator@ycombinator
Doe (@DoeLabs) embeds AI agents into multi-site healthcare groups, starting with DSOs. They plug into a group's PMS, ERP, email, and supplier catalogs to complete tasks, deliver immediate EBITDA lift, and data visibility.

Congrats on the launch, @AdrianBarbir & @YuhuangOu!
14
13/
b-12 is building AI agents that help chemists design experiments and automatically run them on lab robots. They turn months of manual chemistry work into minutes of automated execution. Congrats @drecmb @ZJoncev! @ycombinator


@ycombinator
Y Combinator@ycombinator
b-12 (b12-labs.com) is building AI agents that help chemists design experiments and automatically run them on lab robots. They turn months of manual chemistry work into minutes of automated execution.

Congrats on the launch @drecmb & @ZJoncev!

ycombinator.com/launches/O3V-b…
15
14/
@IronLedgerAI provides AI agents for property accounting, starting with accounts payable. They're already eliminating 100s of hours of work for 1000s of multifamily units every month. @ycombinator


@ycombinator
Y Combinator@ycombinator
IronLedger.ai (@IronLedgerAI) provides AI agents for property accounting, starting with accounts payable.

They're already eliminating 100s of hours of work for 1000s of multifamily units every month.

ycombinator.com/launches/O54-i…
16
15/
@RiselyAI's agents automate admin work across college campuses.

Congrats on the launch, @shahryarsabbasi, @sadiasaifuddin1 & @danial_asif! @ycombinator


@ycombinator
Y Combinator@ycombinator
🚀@RiselyAI's agents automate admin work across college campuses.

Their first product, the AI Advisor, unifies student data to flag at-risk students, deliver personalized support, and improve retention.

Congrats on the launch, @shahryarsabbasi, @sadiasaifuddin1 & @danial_asif!
17
16/
@CyberdeskHQ is the computer use agent for developers to automate legacy Windows apps.

Congrats on the launch, @dope_alan & @mahmoudalmadi_! @ycombinator


@ycombinator
Y Combinator@ycombinator
Cyberdesk (@CyberdeskHQ) is the computer use agent for developers to automate legacy Windows apps. Customers in healthcare, finance, and more use it to automate EHRs and accounting— combining reliable memorized steps with smart fallback during popups.

ycombinator.com/launches/O4a-c…

Congrats on the launch, @dope_alan & @mahmoudalmadi_!
18
17/
PotPie: Open-source platform for custom AI code agents

Spawn faster with less friction. @tom_doerr


@tom_doerr
Tom Dörr@tom_doerr
open-source platform for custom AI code agents
Media image
19
18/
Lovable's new agent writes better code than 90% of human programmers, works 24/7, never quits, and costs less than a Netflix subscription.

@EHuanglu shares 10 apps people built with it.


@EHuanglu
el.cine@EHuanglu
a single AI agent just made 47 million developers obsolete

Lovable's new agent writes better code than 90% of human programmers, works 24/7, never quits, and costs less than a Netflix subscription

10 apps people built in under an hour:

1. voice controlled app with 0 buttons
20
19/
@IshitaJindal17 just shipped 15 plug-and-play templates to build agents:

Browser Agent (scrapes anything)
Hook Generator (viral content)
Reel creator (based on current trends)
+ 12 more


@IshitaJindal17
Ishita Jindal@IshitaJindal17
Stop building AI agents from scratch.

we just shipped 15 plug-and-play templates to build agents:

- Browser Agent (scrapes anything)
- Hook Generator (viral content)
- Reel creator (based on current trends)
+ 12 more

🧵
21
20/
@hwchase17 drops a Python package for building advanced agents with planning tools, sub-agents & file systems. Beats simple loops for complex tasks.


@hwchase17
Harrison Chase@hwchase17
🧠🤖Deep Agents

Simple tool calling loops fail on longer time horizon or more complex tasks

Deep Agents like Deep Research, Claude Code & Manus succeed by using a number of tools and tricks

We created a new Python package which makes it easy to build your own Deep Agents!
Media image
22
21/
Deploy any ML model/RAG/agent as MCP server in 10 lines!

@litserve_ai tutorial: 2x faster than FastAPI, full inference control. @_avichawla


@_avichawla
Avi Chawla@_avichawla
You can now deploy any ML model, RAG, or Agent as an MCP server.

And it takes just 10 lines of code.

Here's a step-by-step breakdown (100% private):
23
22/
@levie explains how agents expand enterprise software TAM by automating scarce work in legal, finance & more. Breaks user constraints for massive growth.


@levie
Aaron Levie@levie
The trillion dollar opportunity in enterprise software is AI Agents. The reason for this is AI Agents expand many software categories because most tools have been constrained by the number of users on the other end of the tool.

Enterprise software traditionally enabled people to do their work. But now the software also comes with actual productive output as well. This then breaks the traditional limits that many software markets have had, because smaller customers can use these tools more, new departments and industries open up, and previously scarce areas of work can be scaled more.

For instance, most estimates put the code IDE category at a few billion dollars just a few years ago. This has - amazingly - always been a very small category of software. Well, now with AI Agents, the IDE market and coding in general is one of the fastest categories of spend in tech. This is because it’s bringing automation to a high value activity and supplying the world with a traditionally very scarce and high cost resource, engineers.

The same is going to be true for many other categories of work across legal, financial services, healthcare, and more. Amazingly, this now opens up many niche categories that just wouldn’t have been economically viable for a software company before. For instance, making a business out of selling software to life sciences regulatory compliance managers would have been sub-scale before, but in a world of AI Agents this all of a sudden becomes possible.

We’re already seeing this in our initial Box AI Agent use cases. Many of the early Agents being created by customers are for automating or augmenting work in previously underserved areas. This is letting customers automate a process they never would have gotten around to before, or dramatically expanding the output of work they were already doing. In all cases it’s more TAM that software wasn’t touching before.

In all, it’s clear that AI Agents are going to grow many software categories. It’s an incredible time to be going after these spaces because even the small ones will now be massive, and the big ones will just get bigger. Tons of opportunity.
Actions
What You Can Do
  • Export as PDF or Markdown
  • Batch Export to Notion
  • Bookmark & Highlight
  • Screenshot Tweet
Create Free Account

Includes 7-day Premium trial

Advertisement