今日科技推文精选 - 2026年4月28日

2026年4月28日科技每日简报

Today's top tech conversations are led by @business, whose post about 'Microsoft will no longer pay r...' garnered the highest engagement. Key themes trending across the top stories include openai, models, https, model, agent. The community is actively discussing recent developments in AI, engineering practices, and startup strategies.

1. business (Group Score: 702.7 | Individual: 52.3)

Cluster: 33 tweets | Engagement: 1125 (Avg: 68) | Type: Tech

Microsoft will no longer pay revenue to OpenAI and said its partnership with the leading artificial intelligence firm will not be exclusive going forward https://t.co/sV4QUPp1i5

See 32 related tweets

@aakashgupta: Microsoft just removed OpenAI's only off-ramp: the AGI clause.

Read the new line. Revenue share con...

@wallstengine: Microsoft and OpenAI rewrote key terms of their partnership:

$MSFT will no longer pay a revenue s...

@rohanpaul_ai: OpenAI is moving away from its exclusive Microsoft arrangement, making room for possible partnership...
@zephyr_z9: The Burry curse\n\nQT @KobeissiLetter: BREAKING: Microsoft stock, $MSFT, falls -5% after announcing ...
@KobeissiLetter: BREAKING: Microsoft stock, $MSFT, falls -5% after announcing that its OpenAI license will now be non...

2. FT (Group Score: 485.4 | Individual: 56.0)

Cluster: 19 tweets | Engagement: 3817 (Avg: 197) | Type: Tech

Breaking news: China has blocked Meta’s $2bn acquisition of artificial intelligence platform Manus, after regulators reviewed whether the deal violated Beijing’s investment rules. https://t.co/hsuAdD1HUB https://t.co/anPdvfcNYJ

See 18 related tweets

@business: China has blocked Meta’s $2 billion acquisition of agentic AI startup Manus, making a surprise move ...
@business: China’s decision to press Meta to unwind a $2 billion acquisition of AI startup Manus marks a step u...
@TFTC21: AI talent is now a national security asset that governments will physically prevent from crossing bo...
@teortaxesTex: This is a somewhat comical situation because the value of Manus is negligible both for the Chinese a...
@cryptopunk7213: i mean what an insane story. meta’s $2B acquisition is now dead:

-> March 2025: Manus launches AI ...

3. XFreeze (Group Score: 249.7 | Individual: 66.5)

Cluster: 7 tweets | Engagement: 15216 (Avg: 970) | Type: Tech

Interesting how it works

Elon puts up his own money, rounds up the absolute best AI talent on the planet, leverages every connection he has to secure serious resources, and launches OpenAI in 2015 as a pure non-profit explicitly created to develop AI for the benefit of humanity, with zero profit motive and open research

Then the “team” decides they want the bag They push Elon out, take control, and quietly flip the entire thing into a for-profit machine

All while preaching the same sanctimonious lines on repeat: “We’re still mission-driven!” “AI for the good of humanity!” “We’d never abandon our principles!”

The ultimate betrayal: Elon got zero equity. Not a single share. He funded it. He built the foundation. He got nothing while they turned his non-profit into their personal cash cow

This is the level of betrayal and hypocrisy we’re dealing with

And for the record.... this lawsuit doesn’t put a single penny in Elon’s pocket. Any win goes straight back to the non-profit to restore the exact mission he founded\n\nQT @OpenAINewsroom: We can't wait to make our case in court where both the truth and the law are on our side. This lawsuit has always been a baseless and jealous bid to derail a competitor. We'll also finally have the chance to question Mr. Musk under oath before a jury of Californians about this attempt to undermine our work to ensure that artificial general intelligence benefits all of humanity.

See 6 related tweets

@elonmusk: Scam Altman\n\nQT @XFreeze: Interesting how it works

Elon puts up his own money, rounds up the abso...

@elonmusk: Exactly\n\nQT @Teslaconomics: Today is the start of a very important day in the AI world... the tria...
@XFreeze: Without Elon Musk, there is no world-class OpenAI team

He recruited the best minds on the planet, p...

@MarioNawfal: 🇺🇸 The 2 most powerful men in AI are about to argue in front of 9 strangers.

@elonmusk v. Altman ju...

@markgadala: The Elon vs OpenAI lawsuit timeline is so insane it's almost impossible for me to believe Elon won't...

4. zephyr_z9 (Group Score: 221.0 | Individual: 53.5)

Cluster: 7 tweets | Engagement: 1202 (Avg: 139) | Type: Tech

RT @sama: we have updated our partnership with microsoft.

microsoft will remain our primary cloud partner, but we are now able to make our products and services available across all clouds.

will continue to provide them with models and products until 2032, and a revenue share through 2030.

See 6 related tweets

@JoshKale: Sam Altman renegotiating with Microsoft:

2019 → "exclusive models, 20% of revenue, deal?" 2023 → "e...

@deredleritt3r: >will continue to provide them with models and products until 2032

Very curious whether there have ...

@TheRealAdamG: Expanded distribution for the win.\n\nQT @sama: we have updated our partnership with microsoft.

mic...

@ShanuMathew93: “we are now able to make our products and services available across all clouds.”\n\nQT @sama: we hav...
@petergostev: Many years too late, but at least nice for the users https://t.co/thFkuLq0Ka\n\nQT @sama: we have up...

5. elonmusk (Group Score: 219.5 | Individual: 42.0)

Cluster: 8 tweets | Engagement: 49594 (Avg: 23131) | Type: Tech

Scam Altman didn’t tell the OpenAI board that he OWNED the OpenAI Startup Fund.

Altman lied in congressional testimony that he didn’t have financial gain from OpenAI.\n\nQT @cb_doge: Ex-board member of OpenAI calls Sam Altman a liar.

He lied to the board for years, hid ChatGPT launch, lied about owning Startup Fund, falsified safety info, and lied to oust her after her paper.

Board lost all trust → fired him.

Sam Altman is a liar. https://t.co/liBDZS0JA9

See 7 related tweets

@cb_doge: Sam Altman cannot be trusted.

• The OpenAI board fired him because he was not always honest with th...

@elonmusk: Scam Altman owned the OpenAI Startup fund while simultaneously lying to the world that he didn’t fin...
@elonmusk: Yup\n\nQT @Z_XSophie: He is “Scam Altman”. Why? Because the pattern is obvious. • Founded OpenAI ...
@elonmusk: Calling him “Scam” Altman is accurate.

This is very much worth reading.\n\nQT @RonanFarrow: The re...

@elonmusk: Long list\n\nQT @XFreeze: Scam Altman has a incredible track record for being a con artist I don't t...

6. hardmaru (Group Score: 197.7 | Individual: 43.5)

Cluster: 6 tweets | Engagement: 331 (Avg: 142) | Type: Tech

For the past few years, humans have been doing “prompt engineering” to coax the best performance out of different LLMs. In this work, we explored what happens if we train an AI to do that job instead.

By training a Conductor model with RL, we found that it naturally learns to write highly effective, custom instructions for a whole pool of other models. It essentially learns to ‘manage’ them in natural language.

What surprised me most was how it dynamically adapts. For simple factual questions, it just queries one model. But for hard coding problems, it autonomously spins up a whole pipeline of planners, coders, and verifiers.

Really excited to see where this paradigm of “AI managing AI” goes next, especially as we start moving from single-agent chain-of-thought to multi-agent “chain-of-command”.

Link to our #ICLR2026 paper: https://t.co/EwbjjRPLUb

Along with our TRINITY paper which we announced earlier, this work also powers our new multi-agent system: Sakana Fugu (https://t.co/2m8VRdOYqG) 🐡\n\nQT @SakanaAILabs: Introducing our new work: “Learning to Orchestrate Agents in Natural Language with the Conductor” accepted at #ICLR2026

https://t.co/Wnh9ZACmLm

What if we trained an AI not to solve problems directly, but to act as a manager that delegates tasks to a diverse team of other AIs?

To solve complex tasks, humans rarely work alone; we form teams, delegate, and communicate. Yet, multi-agent AI systems currently rely heavily on rigid, human-designed workflows or simple routers that just pick a single model. We wanted an AI that could dynamically build its own team.

We trained a 7B Conductor model using Reinforcement Learning to orchestrate a pool of frontier models (including GPT-5, Gemini, Claude, and open-source models available during the period leading up to ICLR 2026).

Instead of executing code, the Conductor outputs a collaborative workflow in natural language. For any given question, the Conductor specifies:

1/ Which agent to call 2/ What specific subtask to give them (acting as an expert prompt engineer) 3/ What previous messages they can see in their context window

Through pure end-to-end reward maximization, amazing behaviors emerged. The Conductor learned to adapt to task difficulty: it 1-shots simple factual questions, but autonomously spins up complex planner-executor-verifier pipelines for hard coding problems.

The results are very promising: The 7B Conductor surpasses the performance of every individual worker model in its pool, setting new records on LiveCodeBench (83.9%) and GPQA-Diamond (87.5%) at the time of publication. It also significantly outperforms expensive multi-agent baselines like Mixture-of-Agents at a fraction of the cost.

One of our favorite features: Recursive Test-Time Scaling! By allowing the Conductor to select itself as a worker, it reads its own team's prior output, realizes if it failed, and spins up a corrective workflow on the fly. This opens a new axis for scaling compute during inference.

This research proves that language models can become elite meta-prompt engineers, dynamically harnessing collective intelligence.

Alongside our TRINITY research which we announced a few days earlier, this foundational research powers our new multi-agent system: Sakana Fugu! (https://t.co/36Ud311KCP) 🐡

OpenReview: https://t.co/e5WqTleQNL (ICLR 2026)

See 5 related tweets

@SakanaAILabs: Introducing our new work: “Learning to Orchestrate Agents in Natural Language with the Conductor” ac...
@WesRoth: Sakana AI unveiled TRINITY at ICLR 2026, challenging the industry's obsession with endlessly scaling...
@hardmaru: RT @SakanaAILabs: Introducing our new work: “Learning to Orchestrate Agents in Natural Language with...
@omarsar0: I consider this one of the most interesting research themes happening in AI today. Worth taking a lo...
@simonguozirui: RT @hardmaru: For the past few years, humans have been doing “prompt engineering” to coax the best p...

7. chatgpt21 (Group Score: 190.4 | Individual: 33.3)

Cluster: 8 tweets | Engagement: 90 (Avg: 180) | Type: Tech

OpenAI is reportedly in development of multiple hardware products including a phone!

The article first speaks about the collab with Jonny Ive “The first product expected to launch is a new HomePod-type smart speaker infused with ChatGPT and built-in camera. Smart glasses and a smart lamp are also reportedly in development.”

While previous reporting has consistently said that OpenAI is not working on a phone, it seems that’s no longer the case.

Ming-Chi Kuo, in a post on X stated that OpenAI is working with MediaTek and Qualcomm to “develop smartphone processors, with Luxshare as the exclusive system co-design and manufacturing partner. Mass production is expected in 2028…Specifications and suppliers are expected to be finalized by late 2026 or 1Q27.”

See 7 related tweets

@synthwavedd: 🚨BREAKING: OpenAI are working on their own phone to compete with the iPhone. Details:
They're wor...
@JoshKale: Wow looks like OpenAI is making a smartphone 👇

→ Mass production 2028. → Luxshare as exclusive sys...

@StockMKTNewz: OpenAI smartphone coming??

OpenAI is reportedly working with MediaTek, Qualcomm $QCOM and LuxShare ...

@wallstengine: Qualcomm $QCOM has now turned negative on the day, giving back all of its premarket gains. https://t...
@kimmonismus: OpenAI is working on a phone.

They are expected to compete with apples iPhone.

First product howev...

8. BrianRoemmele (Group Score: 166.3 | Individual: 62.4)

Cluster: 4 tweets | Engagement: 29924 (Avg: 592) | Type: Tech

RT @elonmusk: Scam Altman and Greg Stockman stole a charity. Full stop.

Greg got tens of billions of stock for himself and Scam got dozens of OpenAI side deals with a piece of the action for himself, Y Combinator style. After this lawsuit, Scam will also be awarded tens of billions in stock directly.

The fundamental question is simply this:

Do you want to set legal precedent in the United States that it is ok to loot a charity? If so, you undermine all charitable giving in the United States forever.

I could have started OpenAI as a for-profit corporation. Instead, I started it, funded it, recruited critical talent and taught them everything I know about how to make a startup successful FOR THE PUBLIC GOOD.

Then they stole the charity.

See 3 related tweets

@niccruzpatane: "Non-Profit" but is driving a $2M+ Hypercar with the windows down

Make it make sense https://t.co/q...

@Hadas_Gold: Musk has pinned this post on the day his trial v Altman begins\n\nQT @elonmusk: Scam Altman and Greg...
@ns123abc: 🚨 BREAKING: Elon Musk just called out OpenAI co-founders while his jury is being selected:

“Scam ...

9. MikeIsaac (Group Score: 152.6 | Individual: 31.4)

Cluster: 6 tweets | Engagement: 93 (Avg: 2105) | Type: Tech

one of the big advantages Anthropic had in its Claude Code enterprise growth was having its models available on AWS, Google cloud and Msft azure, etc

for a long time, OAI was primarily available via azure

enterprise clients often didnt want to switch if they were on AWS or GC\n\nQT @ajassy: Very interesting announcement from OpenAI this morning. We’re excited to make OpenAI's models available directly to customers on Bedrock in the coming weeks, alongside the upcoming Stateful Runtime Environment. With this, builders will have even more choice to pick the right model for the right job. More details at our AWS event in San Francisco tomorrow.

See 5 related tweets

@TheRealAdamG: OpenAI on AWS?!

I hope that doesn't cause a "constitutional crisis" for any of our competitors....

@Jessicalessin: Take note when anyone calls an announcement "very interesting."\n\nQT @ajassy: Very interesting anno...
@TheRealAdamG: RT @ajassy: Very interesting announcement from OpenAI this morning. We’re excited to make OpenAI's m...
@wallstengine: Amazon’s $AMZN Andy Jassy called OpenAI’s announcement this morning “very interesting” and said AWS ...
@StockSavvyShay: $AMZN CEO Andy Jassy called OpenAI’s announcement “very interesting” and said AWS plans to add OpenA...

10. rohanpaul_ai (Group Score: 145.1 | Individual: 34.7)

Cluster: 6 tweets | Engagement: 41 (Avg: 54) | Type: Tech

OpenAI is reportedly exploring a smartphone that treats AI agents as the operating model, not apps.

For this, as per news, OpenAI is exploring custom mobile silicon with MediaTek, Qualcomm, and Luxshare, in a co-design and manufacturing role, because agentic computing is not just model inference, it is power efficiency, memory management, and deciding what stays on-device versus what goes to the cloud.

Aiming to replace much of the app model with a system that acts across the whole device.

The deeper logic is vertical integration. OpenAI has already folded Jony Ive’s io team into the company, and Chris Lehane has said its first hardware product is on track to be announced in the second half of 2026, which suggests a long game around owning more of the stack, not just shipping another companion app.

The basic idea is that apps split your tasks into separate boxes, while an agent-first phone would watch context across messages, location, voice, calendar, and settings, then complete actions directly instead of handing you 10 separate screens.

That matters for capability because Apple and Google decide what normal apps can access, so a full hardware and software stack could let OpenAI connect models more deeply to calling, writing, booking, search, reminders, and device controls.

The likely technical setup is a split between small on-device models for fast, private, low-latency work and cloud models for heavier reasoning, which is the practical way to make an always-available assistant feel responsive without draining the phone.

A phone is the densest stream of personal context most consumers generate, so a true AI agent phone would not merely answer questions better, it would see more, remember more, and infer more than an app sitting inside Apple’s or Google’s rules can.

techcrunch .com/2026/04/27/openai-could-be-making-a-phone-with-ai-agents-replacing-apps/

See 5 related tweets

@cryptopunk7213: oh shit openai’s building a smartphone device to compete with Apple

this makes a lot of sense… if t...

@ianmiles: Yeah I would never get this. Imagine trusting Sam Altman with your phone.\n\nQT @Polymarket: JUST IN...
@wallstengine: Ming-Chi Kuo reports that OpenAI is working with MediaTek, Qualcomm $QCOM and Luxshare on an AI-firs...
@StockSavvyShay: OpenAI is reportedly working with $QCOM on an AI-first smartphone targeted for 2028 mass production....
@TechCrunch: OpenAI could be making a phone with AI agents replacing apps https://t.co/M79t6fh7Qv...

11. Reuters (Group Score: 142.4 | Individual: 32.3)

Cluster: 6 tweets | Engagement: 125 (Avg: 98) | Type: Tech

Market reaction to DeepSeek's preview of its long-awaited next-generation artificial intelligence model has so far been subdued compared with the Chinese startup's outsized global breakthrough last year after the launch of its low-cost AI models https://t.co/AAMpBFjesV

See 5 related tweets

@business: DeepSeek is aggressively pitching low-priced-plans for its just-released flagship model, intensifyin...
@Reuters: China's DeepSeek slashes prices for new AI model https://t.co/9HYhSHjFHQ https://t.co/9HYhSHjFHQ...
@dee_bosa: Watch cloud revenue growth at Amazon, Google and Microsoft this week as they all report on Weds

It...

@SarithaRai: Price Drop -- DeepSeek is pitching its newest model DeepSeek-V4 at aggressive prices, accelerating t...
@Reuters: DeepSeek's new AI model does not wow markets in fast-changing industry https://t.co/WBXktdglBN https...

12. github (Group Score: 140.2 | Individual: 35.7)

Cluster: 6 tweets | Engagement: 4009 (Avg: 805) | Type: Tech

Starting June 1st, GitHub Copilot will move to a usage-based billing model as GitHub Copilot supports more agentic and advanced workflows.

In early May, you'll see a preview bill experience, giving visibility into projected costs before the transition.

👉 Read more about the upcoming change: https://t.co/4IC9VNHwhk

See 5 related tweets

@GergelyOrosz: Its the beginning of the end of subsidized AI subscriptions. GH Copilot is moving to usage-based bil...
@scaling01: "GitHub Copilot will move to a usage-based billing" on June 1, 2026

now let's see AI demand after i...

@edzitron: https://t.co/klfoAUeP62 https://t.co/sohbgNi4Vj\n\nQT @github: Starting June 1st, GitHub Copilot wil...
@edzitron: https://t.co/fp4Epu3T61\n\nQT @Techmeme: GitHub says all Copilot plans will move to usage-based bill...
@Techmeme: GitHub says all Copilot plans will move to usage-based billing on June 1, replacing premium requests...

13. wallstengine (Group Score: 133.0 | Individual: 26.5)

Cluster: 9 tweets | Engagement: 81 (Avg: 135) | Type: Tech

More than 580 Google employees, including AI researchers and some senior leaders, signed a letter urging Sundar Pichai to reject classified U.S. defense AI work, warning the systems can make mistakes and could be used in harmful ways without oversight. https://t.co/3lCdKAfAp0

See 8 related tweets

@FT: More than 560 Google employees have signed an open letter to chief executive Sundar Pichai urging hi...
@Polymarket: JUST IN: 560+ Google employees sign letter urging CEO Pichai to reject military use of Google AI, wa...
@business: Hundreds of AI researchers at Google have signed a letter urging CEO Sundar Pichai not to allow the ...
@Techmeme: More than 600 Google employees, including many from DeepMind, sign a letter to Sundar Pichai demandi...
@KatrinaManson: RT @byJuliaLove: 🚨New scoop from me and @KatrinaManson: More than 580 Google workers including hundr...

14. wallstengine (Group Score: 129.3 | Individual: 34.2)

Cluster: 5 tweets | Engagement: 195 (Avg: 135) | Type: Tech

$META has agreed to buy up to 1 gigawatt of space-based solar power from Overview Energy to help run its AI data centers, with commercial delivery expected in 2030. The startup plans its first orbital demo in 2028. https://t.co/vGU51nfiyR

See 4 related tweets

@StockSavvyShay: $META agreed to buy up to 1 gigawatt of space-based solar power from Overview Energy for its AI data...
@StockMKTNewz: Meta Platforms $META just signed a purchase agreement to get

SOLAR POWER FROM SPACE AT NIGHT

Meta...

@business: Meta is looking to power AI data centers with solar energy collected in space, taking a novel approa...
@Techmeme: Meta says it signed a deal with Overview Energy for 1GW of space solar energy; the startup seeks to ...

15. TheAhmadOsman (Group Score: 127.0 | Individual: 35.0)

Cluster: 4 tweets | Engagement: 329 (Avg: 239) | Type: Tech

New Opensource SoTA contender enters the arena

Xiaomi MiMo-V2.5 Pro

1.02T Total Params / 42B Active Params
Base and Instruct versions

Xiaomi MiMo-V2.5

310B Total Params / 15B Active Params
Base and Instruct versions

MIT License

Opensource AI just keeps getting better https://t.co/pTJKNXvA0i\n\nQT @XiaomiMiMo: Xiaomi MiMo-V2.5 is now officially open-sourced！ MIT License, supporting commercial deployment, continued training, and fine-tuning - no additional authorization required. Two models, both supporting a 1M-token context window : • MiMo-V2.5-Pro: built for complex agent and coding tasks, ranking No.1 among open-source models on GDPVal-AA and ClawEval • MiMo-V2.5: a native omni-modal model with strong agent capabilities A model's value isn't measured by rankings alone — it's measured by the problems it solves. Let's build with MiMo now! 🤗 Weights: https://t.co/w7wNtXj9Rm 📄 Blog: https://t.co/bPaaWMRI0g

See 3 related tweets

@vllm_project: 🎉 Day-0 vLLM support for the MiMo-V2.5 series! Congrats to @XiaomiMiMo on the open-source release of...
@scaling01: btw that's another 1T open-source model\n\nQT @XiaomiMiMo: Xiaomi MiMo-V2.5 is now officially open-s...
@QuixiAI: Awesome! @XiaomiMiMo v2.5 is 33% Smaller than DeepSeek and multimodal! A true competitor for Kimi K...

16. codewithimanshu (Group Score: 126.7 | Individual: 34.0)

Cluster: 4 tweets | Engagement: 350 (Avg: 764) | Type: Tech

Anthropic's Applied AI team just dropped a 25-minute talk on how to actually use Claude Code properly.

Free. From the people who built it.

You've been using Claude Code for months without the workflows they teach in this.

The official best practices no $500 course covers. Full breakdown below.

Bookmark it.

Follow @codewithimanshu for more high-signal content that actually moves your skills forward.

↓

Cal Rueb runs Applied AI at Anthropic. He just gave away the official Claude Code playbook.

This is the talk that separates people shipping production apps with Claude from people stuck debugging the same errors for hours.

If you're using Claude Code without watching this, you're leaving 80% of its power on the table.

Follow @codewithimanshu for more Claude Code breakdowns weekly.

↓

The architecture nobody explains.

Claude Code isn't a chatbot with code generation.

It's a pure agent. An autonomous loop of instructions and tools that runs until a task is complete.

It doesn't pre-index your codebase. It uses agentic search in real time. Grep, find, glob. Exploring your files like a senior engineer onboarding to a new repo.

This architecture is why Claude Code outperforms every other coding assistant. Most people don't even know it works this way.

Follow @codewithimanshu for production AI agent architecture content.

↓

The 4 use cases Claude Code is built for.

Onboarding and discovery on a new codebase. Planning new features as a thought partner. Building from scratch or contributing to existing projects. Managing complex tasks like full codebase migrations.

If you're using Claude Code for anything outside this, you're using it wrong.

Cal explains exactly when to lean on each mode and how to switch between them.

Follow @codewithimanshu for daily Claude Code workflows that actually ship.

↓

CLI integration is where Claude Code becomes superhuman.

Git rebasing without breaking your repo. Docker configurations without 50 Stack Overflow tabs. Complex shell scripting on the fly.

Claude Code handles CLI tools natively. You describe what you want. It executes the right commands.

Most people use Claude Code without ever connecting it to their CLI workflow. Then wonder why they're still slow.

Follow @codewithimanshu for the CLI patterns that 10x Claude Code workflows.

↓

The claude md file is your unfair advantage.

This is the file that separates production users from hobby users.

Project-specific instructions. Style guides. Documentation. Architecture decisions.

Set it up once and Claude Code stops making the same mistakes every session. Stops asking for context you've already given. Stops writing code that violates your conventions.

I'll post my exact claude md template next week.

Follow @codewithimanshu so you don't miss it.

↓

Context management for long sessions.

Claude Code runs on a 200,000-token context window. Massive. But not infinite.

When you approach the limit, use /compact to summarize progress and continue without losing momentum.

Most people hit the limit, get confused responses, and blame the model.

The model is fine. Their context management isn't.

Follow @codewithimanshu for context management patterns that prevent production AI failures.

↓

Permission management for speed.

Auto-accept mode for routine commands. Manual approval for destructive operations.

Configure it once and Claude Code stops asking for permission on every single file edit.

Your workflow goes from "click yes 50 times an hour" to "Claude executes while you think."

This single setting saves hours per week.

Follow @codewithimanshu for the exact permission setups that production AI teams use.

↓

Iterative workflows that actually ship.

Test-driven development. Commit frequently. Use think hard for deeper reasoning.

Cal's framework: write the test first. Let Claude build to pass it. Commit when it does. Roll back when it doesn't.

This is how you avoid the nightmare scenario where Claude has been working for 30 minutes and produced 2,000 lines of code you can't merge.

The think hard command alone changes output quality dramatically.

Follow @codewithimanshu for daily Claude Code prompts that ship code.

↓

Multi-agent orchestration.

Run multiple instances of Claude Code in parallel. Different terminals, different tasks.

One agent refactoring the backend. Another writing tests. A third building the frontend.

You're orchestrating an entire engineering team from your laptop.

Most people run one Claude Code session at a time and call it productivity. The serious users run 4 in parallel.

Follow @codewithimanshu for multi-agent workflows that scale solo developers into entire teams.

↓

Emergency controls every user needs.

Escape to stop the agent when it heads in the wrong direction. Escape twice to navigate back through conversation history.

Most people frantically Ctrl+C when Claude goes off track. Escape is cleaner. Recovers context. Saves your session.

Small detail. Massive workflow improvement.

Follow @codewithimanshu for daily Claude Code shortcuts.

↓

MCP servers extend everything.

Model Context Protocol servers let Claude Code interact with anything beyond your terminal.

Databases. APIs. Internal tools. Custom workflows. Vendor integrations.

This is where Claude Code stops being a coding assistant and becomes infrastructure.

The teams shipping AI products at scale all use MCP. Solo devs are mostly still ignoring it.

Follow @codewithimanshu for MCP server setups that turn Claude Code into a full-stack engineering team.

↓

What you'll have after watching this.

Claude Code configured the way Anthropic engineers configure it. Workflows that ship code instead of just generating it. Context management that survives 8-hour sessions. Multi-agent orchestration running parallel tasks. The exact best practices the team building it uses internally.

That's not "knowing how to use Claude Code." That's mastery.

The kind of mastery that turns into $200K AI engineering roles and$ 10K/month consulting contracts.

↓

25 minutes from the team that built it.

You'll spend longer in meetings tomorrow and learn nothing.

This compounds for every line of code you write for the rest of your career.

People who watch it ship 10x faster with Claude Code. People who skip it keep getting frustrated wondering why everyone else is shipping faster than them.

Save the video. Watch it tonight. Apply the practices tomorrow.

Follow @codewithimanshu for more high-signal Claude Code content that actually moves your engineering career forward.

See 3 related tweets

@aakashgupta: Claude Design just split PMs into two categories. The ones running the loop ship 500 prototype reps ...
@codewithimanshu: RT @codewithimanshu: Anthropic's Claude Ai Agents Team just Educated how to build production AI agen...
@codewithimanshu: RT @codewithimanshu: Anthropic's Claude Code team just teaches how to automate your entire engineeri...

17. WesRoth (Group Score: 123.3 | Individual: 33.0)

Cluster: 4 tweets | Engagement: 44 (Avg: 25) | Type: Tech

Browser Use launched "Browser Use Box (bux)," a dedicated, cloud-based environment that hosts a 24/7 personal AI agent powered by Browser Harness.

This system directly solves the limitation of local AI tools that terminate the moment a laptop is closed.

"Bux" runs continuously on a server, utilizing a real Chrome browser with persistent logins, and integrates seamlessly with Telegram so users can text commands to their agent from anywhere.\n\nQT @larsencc: Introducing: Browser Use Box (bux). Your 24/7 personal agent box, powered by Browser Harness. ♞

We got tired of agents that vanish when you close the laptop. So we put them on a server.

24/7 box that runs while you sleep Real Chrome with persistent logins Telegram baked in, text from anywhere

I gave mine my email and a Telegram chat. It books flights, replies to LinkedIn and does my to-do list before I'm awake.

Try it at https://t.co/kABDn9Q6AL ↓

See 3 related tweets

@larsencc: Bux exists because @mathisdittrich kept pushing for it.

Six of us at @browser_use had the same kind...

@garrytan: This is actually very awesome\n\nQT @larsencc: Introducing: Browser Use Box (bux). Your 24/7 persona...
@larsencc: First 24h of bux.

The #1 ask was the same thing we wanted internally: > scheduling > queuei...

18. elonmusk (Group Score: 107.7 | Individual: 38.2)

Cluster: 3 tweets | Engagement: 34064 (Avg: 23131) | Type: Tech

This is why OpenAI was created\n\nQT @cb_doge: Elon Musk turned down all shares when he left OpenAI because he believed nonprofits are not meant for self-enrichment.

"The reason I founded OpenAI was because I was concerned, based on my conversations with Larry Page, that he was not sufficiently concerned about the dangers of AI. At my birthday party, he, in front of a large group of people, called me a speciesist, for favoring humanity over computers. So after that, I was like, We got to have some counterbalance to Google, because Larry doesn't seem to care if humans make it or not.

So I thought, what's the opposite of Google? It would be an open source nonprofit, and that's where the word open, in OpenAI comes from. It means open source.

I provided all the money, recruited the key people, and taught them everything I know. I actually even got them to deal with Microsoft.

And for all that, I did not seek any financial reward whatsoever. The reason I actually took down the offer for shares is because, I mean, I felt like what are the shares, and why like nonprofits supposed to have shares? Nonprofits are not supposed to be self enrichment, so that's why I turned on the offer of shares."

— Elon Musk

See 2 related tweets

@cb_doge: Elon Musk turned down all shares when he left OpenAI because he believed nonprofits are not meant fo...
@XFreeze: Elon Musk is the reason OpenAI exists.... it wouldn't have existed without him

He built the vision....

19. petergyang (Group Score: 102.1 | Individual: 42.0)

Cluster: 3 tweets | Engagement: 177 (Avg: 92) | Type: Tech

My top 5 takeaways from @tibo_maker, a solo AI founder who's making $1M+ a month:

Charge money on day one.

Tibo’s first startup failed because he cared more about appearing successful (e.g., I managed a team of 10 and raised $200K) than validating demand with paying customers. “If there is no revenue and no stickiness in the revenue, it’s going to be very hard to build a successful business.” Free signups are easy to mistake for traction.

Follow the signal when users surprise you.

Tibo acquired Typeframe ( $2K MRR) as a product video tool, but noticed users were hacking it to stitch 5-second AI clips into longer videos with consistent characters and scenes. He pivoted the entire product to meet this need and rebranded it to Revid, which is now making$ 600K+ MRR.

Price your AI SaaS at $50-100/month

Low enough that customers don’t need a sales call and high enough to filter out tire-kickers. “I see so many people charging $10 / month and it puts you into the position of a cheap product.” Tibo picks his price point first, then shapes the product around it.

Keep monthly churn below 20%.

If more than 20% of customers cancel each month, stop scaling acquisition and fix the product first. There’s a ceiling (max MRR) on your revenue based on churn vs. acquisition. At 40% churn, customers stay about 2 months and you’ll hit a wall no matter how much you spend.

Build tool pages to rank on Google

Revid has 100+ pages each targeting a specific Google search like “turn audio into video” and “YouTube to shorts.” Many AI founders follow a similar model.

📌 Watch our full conversation for more practical tactics like the above: https://t.co/N6b950Xc5p\n\nQT @petergyang: "I shipped 9 failed products before one took off...now I'm doing $1M+/month."

Here's my new episode with @tibo_maker, a solo founder who bootstrapped 5 AI products to $1M+ / month.

Tibo walked me through his exact playbook:

✅ How to validate ideas and fail fast ✅ Why his top acquisition channel is still SEO ✅ The pricing sweet spot for AI products

Some quotes from Tibo:

"When people twist your product into something else, that's a very strong signal you have to follow."

"It's easy to lie to yourself [with free users], but if there's no stickiness in the revenue, it's very hard to build a successful business."

"I'm convinced right now that just one person can do the job of 20 people."

📌 Watch now: https://t.co/N6b950Xc5p

Thanks to our sponsors:

@WisprFlow: Don't type, just speak https://t.co/oqHJ8bN3ll

@linear: The AI agent platform for modern teams https://t.co/tgWf9oL4bs

See 2 related tweets

@petergyang: RT @petergyang: "I shipped 9 failed products before one took off...now I'm doing $1M+/month."

Here'...

@tibo_maker: RT @petergyang: My top 5 takeaways from @tibo_maker, a solo AI founder who's making $1M+ a month:

1...

20. TeksEdge (Group Score: 99.6 | Individual: 31.4)

Cluster: 4 tweets | Engagement: 29 (Avg: 14) | Type: Tech

🚨 New Model Alert: @TencentHunyuan has formally announced Hy3 Preview 🔥

This is China’s newest open-source model (China has been shipping):

🧠 295B MoE (only 21B active) 📏 256K context window ⚡ 40% inference efficiency gain 🛠️ Built for agents, coding & deep reasoning ✅ Fully open-sourced (weights + code) 📅 Released April 23, 2026

Already live across Tencent products and FREE on @OpenRouter

The open-source frontier is pushing out again.\n\nQT @TencentGlobal: Meet Hy3 preview: @TencentHunyuan's most capable Hy model yet.

Built for coding, search, agents, and next-gen applications. 256K context window. 40% inference efficiency gain. Open-source and available now.

📖 https://t.co/3cytnXfsGW

#AI #LLM #AIForGood https://t.co/C7kGtZb1T1

See 3 related tweets

@ivanfioravanti: Tencent Hunyuan Hy3 Preview is here! 🚀 Download in progress! https://t.co/uZULkO9stq\n\nQT @TencentG...
@TeksEdge: 💡 Hy3 is a really capable new open source model from @TencentHunyuan (🇨🇳). It beats Kimi K2.5 and GL...
@mervenoyann: RT @TencentHunyuan: 👋Hi /haɪ/, we're the Tencent Hy /haɪ/ team🐧

Today, we open source Hy3 preview (...