Arindam Majumder 𝕏 (@arindam_1729) — sponsored posts & rates

unfair.so

Browse more creators

@arindam_1729

Arindam Majumder 𝕏8.6K

Arindam Majumder 𝕏

This creator added themselves. Not verified by unfair.so.

@arindam_1729AI AgentsAI CodingOpen Source AI

a builder. founder @studio1hq youtube: https://t.co/ZE1x8oT6Ac

8.6KFollowers

929Median impr. / post

0.7%Engagement

Paid partnerships

In their audience

14 from @ycombinator follow Arindam Majumder 𝕏

@lutherlowe

@ishaansehgal

@_devp

@RosinolToni

@suryaoruganti

@RespanAI+8

6 affiliated with @copilotkit

5 affiliated with @appwrite

5 affiliated with @nebiusai

4 affiliated with @freecodecamp

4 affiliated with @google

1,051 verified followers analyzed

@lutherlowe

@ishaansehgal

@_devp

@RosinolToni

@suryaoruganti

@RespanAI+1045107 AI engineers104 AI founders75 engineers67 founders60 devtools engineers51 AI creators

25–34 core age group

41.2%

31.1% of audience in India

🇮🇳

Aug 31 – Sep 692 posts analyzed · replies not counted · updates weekly

ImpressionsEngagements

9.8K4.9K0

129650

Aug 31Sep 1Sep 2Sep 3Sep 4Sep 5Sep 6

Past collabs

Arindam Majumder 𝕏@arindam_1729 · Jul 31I tested DeepSeek V4 Flash by asking it to build a 3D Snake game. It got to a solid result in just 3 iterations. Total usage: > ~2.8M input tokens > 8.2k output tokens > Under $1 for the entire task That's a pretty impressive price-to-quality ratio. Just a day after OpenAI's pricing update, DeepSeek drops this. Competition in AI models is getting wild. Here's a quick demo of what it built 👇

NebiusPaid partnership8.1K views

Arindam Majumder 𝕏@arindam_1729 · Jul 27Just tested @Kimi_Moonshot K3 against GPT-5.6 and Grok 4.5 on the same task: build a Subway Surfers game. Kimi K3 was my favorite. Better UI, smoother gameplay, and the overall experience felt more polished. GPT-5.6 was solid too, while Grok 4.5 fell behind on the UI. The interesting part is that all three needed at least one follow-up prompt. Kimi also took a bit longer to generate, but the extra time paid off. Ran Kimi K3 through @nebiustf. Low inference cost + a really capable open source model is a great combo.

NebiusPaid partnership46.6K views

Arindam Majumder 𝕏@arindam_1729 · Jul 21Most AI cost optimizers ask you to trade quality for price. Ship claims you don't have to. Just change: `model="<original>"` to: `model="ship-like/<original>"` Same reference model as the quality target. 50% lower price. The interesting idea isn't cheaper inference. It's treating the model name as a specification, then figuring out the cheapest way to meet it at runtime. That's a much bigger shift than model routing.

WithmartianPaid partnership546 views

Arindam Majumder 𝕏@arindam_1729 · Jul 14Was planning a family dinner and didn't want to jump between Google Maps, review sites, and restaurant apps. So I texted @airtap_ai on iMessage: > "What are the best dinner spots nearby?" Then: > "Book a table at one of the restaurants for me" Airtap searched, compared options, sent back the top 3 restaurants, then opened a booking window. I never had to leave the conversation. You're not just chatting with an AI. You're texting your apps through it. No app download. Just send a text to +1-650-213-7322 #justtextit #textyourapps #airtap

AirtapPaid partnership934 views

Arindam Majumder 𝕏@arindam_1729 · Jul 3This is exactly what I was talking about 👀 @CopilotKit just open-sourced Open Generative UI, bringing Claude Artifacts-style experiences to any app. Instead of picking from pre-built components, the agent generates the UI from scratch and streams it into your app in real time. Built on AG-UI, so it works with: - LangGraph, - CrewAI, - Mastra, - Google ADK, - AWS Strands, and more. AI-native interfaces are getting very real.

CopilotkitPaid partnership3K views

Arindam Majumder 𝕏@arindam_1729 · Jun 29I've been saying this for a while: The model isn't the whole coding experience. The harness matters just as much. How the agent explores your codebase, uses tools, recovers from errors, and keeps context often has a bigger impact than swapping one strong model for another. Also, trying every new open-weight model usually means another provider, another API key, and another set of rate limits. ClinePass by @cline is a nice way to simplify that while still letting you bring your own keys or run local models if you want. I recorded a quick walkthrough below 👇

ClinePaid partnership2.1K views

Arindam Majumder 𝕏@arindam_1729 · Jun 16Just tried @Zai_org GLM-5.2 on @nebiustf playground. The best Chinese open-weight model is now accessible through a simple API. - $1.40 input. - $4.40 output per million tokens. - 1M token context window. For comparison: frontier models charge 4-5x more for the same context. Gave it a reasoning prompt: • 559ms time to first token • 27.1s total time • 93.7 tokens/s One catch. Speeds vary as load increases. But for an open-weight model with this much context, performance holds up. Worth testing if you're building reasoning workflows. Demo:

NebiusPaid partnership8.8K views

Arindam Majumder 𝕏@arindam_1729 · Mar 9You can use Claude Code to build and deploy full-stack apps with real infrastructure. Not just frontend code. Actual backend with database, auth, storage, and live deployment. I used @insforge MCP to build a Reddit clone. One prompt: Communities, posts, comments, nested replies, upvoting, image uploads, and deployed with a live link. Claude Code handled the database schema, authentication, file storage, and deployment. Everything. This is what agent-native backends look like. Here's the demo:

InsforgePaid partnership3.3K views

Arindam Majumder 𝕏@arindam_1729 · Feb 24AI agents can build beautiful frontends in minutes. But they hit a wall at backend config. Firebase auth? 15+ manual steps through dashboards. Supabase MCP? Fragmented context that causes hallucinations. @insforge fixes this 👀 Open-source backend built specifically for AI agents: • Full database context in one call • Zero-config auth (Google, GitHub, etc) • 2x more accurate than Supabase MCP • 1.6x faster execution • 30% fewer tokens I built a complete CRM with Claude Code using InsForge. Auth, database, storage, AI integration. Zero manual setup. Works with Cursor, Windsurf, Claude Code, Codex. Repo in replies 🔗

InsforgePaid partnership6K views

Arindam Majumder 𝕏@arindam_1729 · Feb 6Organizing my Mac desktop is always a pain 😤 But why even do it myself when AI can? Tried it with @MiniMax_AI Agent Desktop, and it literally cleaned up and organized my entire desktop for me. Zero effort. Actually useful automation 👀

MinimaxPaid partnership1.6K views

Arindam Majumder 𝕏@arindam_1729 · Feb 5As a DevRel, making presentations is always a pain 😅 I’ve tried a bunch of tools to automate this, but most feel half-baked. Recently gave @MiniMax_AI Agent Desktop’s presentation generator a shot. Dropped in a single link… • Slides were structured well • Content actually made sense • Aesthetics were on point This one surprised me! Here's how I used it:

MinimaxPaid partnership1.3K views

Arindam Majumder 𝕏@arindam_1729 · Dec 15Recently, vibe-coded a real-time urgent care availability that looks for the earliest appointment across 10 clinics in seconds. Handles different booking systems, navigates forms with names and details, and extracts available appointment slots. No APIs needed. Just URLs and what you want.

TinyfishPaid partnership4.2K views

Arindam Majumder 𝕏@arindam_1729 · Sep 27Tried out @CodeRabbit CLI today 🐇⚡ And honestly… I’m impressed. > The file tree view makes the DX great! > It spotted potential issues in my code. > Even explained + suggested AI agent prompts to fix them Overall, big fan of this workflow 👀

CoderabbitPaid partnership1.1K views

Planning a launch?Arindam Majumder 𝕏 is one of 200 curated B2B creators. Tell us what you are launching and we will put the shortlist together.

[email protected]·t.me/fdrtrm

1 picked$100