Prabhat

Building Agents !!

[Prabhat @dead_agent]

@coryalthoff $100 - 5X
And I'm nobody.
Useless post. Just farming impressions !!

[Prabhat @dead_agent]

A country that bans Huawei and pressures companies to stop supplying it hardware now wants to lecture India?
Elon Musk, bro, spare us the gyan. Without your PR machine, your 20x overvalued empire would’ve faced reality long ago.
Look at how the US treats others before preaching to India. #PresidentMusk #Ambani

[Prabhat @dead_agent]

@shreyanshexists Mostly custom. We r built the evals around our actual agent workflows rather than relying purely on generic benchmarks

[Prabhat @dead_agent]

Tomorrow I’m running all our evals on the new Haiku 5.5 and benchmarking it against Luna. If the results hold up, I’ll shift some of our agents to Haiku.

The benchmark is honestly insane. At this pace, the future looks brutal for open-source models and especially for the Chinese model ecosystem.

[Claude @claudeai]

Haiku 5.5 is a significant step up over Haiku 4.5 across coding, computer use, and knowledge work.

[Prabhat @dead_agent]

@shreyanshexists These are our internal agent evals , mainly focused on tool calling, reasoning, accuracy, instruction following, latency, and reliability across real finance workflows.

I’m running the same eval suite across Haiku 5.5 and Luna so it’s an apples-to-apples comparison.

[Prabhat @dead_agent]

I can talk about 10 different things because I’ve worked across multiple modern tech stacks. More importantly, I understand how companies are actually hiring forward looking AI engineers and how customers are adopting AI in real businesses.

So don’t just copy-paste useless posts and spread made-up information for the sake of getting reach.

[Prabhat @dead_agent]

@DrAditya2935 He died 10 years ago as well !!

[Prabhat @dead_agent]

I don’t think @OpenAI has tested #dots properly yet.
It has to be tested properly.

[Prabhat @dead_agent]

@savesage_club
You guys completely fucked up the UI. I’ve paid for your subscriptions and loved the app, but this new UI is a massive downgrade. Feels third-class.

[Prabhat @dead_agent]

Rever Agents !!

[Rever @Reveragents]

Rever is live.

AI agents that do the checking in finance: reconcile, verify, explain the difference, and draft the entry.

A person approves before anything posts.

https://t.co/qjiGx0YuiM

[Prabhat @dead_agent]

@hollllyli You handled it well !!

[Prabhat @dead_agent]

@seyiProducts https://t.co/VJlMVXA3ZC
loved yours !!

[Prabhat @dead_agent]

@join_arc https://t.co/uZfxunx2G5
loved your product !!

[Prabhat @dead_agent]

@keviduk @profound https://t.co/VJlMVXA3ZC

[Prabhat @dead_agent]

🚀 Building the future of Finance Agents.

I’ve been working on the UI for our Finance Agent platform - bringing AI agents into the workflows where finance teams actually work.

The goal isn’t just to build another AI chat interface.

It’s to create a system where specialized agents can work across finance operations - understanding context, connecting to financial systems, and helping teams get things done.

From AP & AR to reporting, reconciliation, accounting and financial data, the vision is to make finance workflows more intelligent and increasingly agent-driven.

I’ve put together the first version of the UI here 👇

🔗 https://t.co/OEZaq7dqcK

Still early. Still building. But this is getting exciting.

AI writes code.
Finance Agents will increasingly do the work. ⚡
#AI #FinanceAI #FinanceAgents #AIAgents #FinTech #Accounting #SaaS #FutureOfWork #BuildInPublic

[wei-wei @wuweiweiwu]

Introducing Mo, the world's first AI QA engineer

Point it at your app and 100+ agents bug bash it in parallel. Every bug comes back with repro steps, logs, and a video

In our evals vs Codex with Playwright MCP: → 3.8x more bugs → 9x more bugs per hour → half the cost per bug

RT + reply "Mo" for $250 in free credits. Drop your app URL, Mo will bug bash it and DM you the report today

[Prabhat @dead_agent]

@anulagarwal Got 750

[Prabhat @dead_agent]

Fucking Insane !!
Claude Opus 5.5