Arbaz Ali | AI Engineer

🧠 AI Agent Engineer | LangChain • LangGraph • RAG • MCP | Building next-gen AI tools | Python + FastAPI 🚀

[Arbaz Ali | AI Engineer @ArbazAliML]

OpenAI vs Anthropic (2026):

OpenAI wins for everyday users with web search, images, memory, and an all-in-one experience.

Anthropic wins for developers with MCP, Claude Code, and complex tasks.

OpenAI powers everyone. Anthropic empowers builders.

#AI #OpenAI #Anthropic

[Anthropic @AnthropicAI]

We’re launching Claude Corps, a national fellowship program matching people early in their careers with US nonprofits.

We'll teach 1,000 people to use Claude, and pay them to use AI to advance their hosts’ missions.

https://t.co/QI6JmlAdSr

[Arbaz Ali | AI Engineer @ArbazAliML]

@jpschroeder @reach_vb 2.3M downloads per week is impressive. Well deserved—excited to see what you build with Codex for Open Source 🚀

[Arbaz Ali | AI Engineer @ArbazAliML]

@gui_penedo @HKydlicek Thanks for the reply! Wishing the team the best with the launch 🚀

[Arbaz Ali | AI Engineer @ArbazAliML]

@gui_penedo @HKydlicek Exciting launch! FineWeb datasets have already become a go-to reference in the LLM pre-training space. Curious to see how Macrodata Labs pushes data quality even further — great data truly is the foundation everything else builds on. Following closely! 🚀

[Arbaz Ali | AI Engineer @ArbazAliML]

A smarter LLM won't fix bad retrieval.

RAG uses semantic vector search to find relevant context. Hybrid RAG combines vector and keyword/BM25 search, capturing both meaning and exact matches.

Result: better recall, higher accuracy, and fewer hallucinations.

[Arbaz Ali | AI Engineer @ArbazAliML]

@sama The electricity analogy is cute. Electricity also took 40 years to reach everyone. Can't wait for my grandkids to afford the everyone plan.

[Arbaz Ali | AI Engineer @ArbazAliML]

@reach_vb The shift isn't from coding to prompting.

It's from giving instructions to defining success.

"Make it faster" is a task.

"Reduce build time, verify it, and show the results" is an objective.

That's where Codex excels.

[Arbaz Ali | AI Engineer @ArbazAliML]

@reach_vb @reach_vb Congrats on 50K! 🎉 Following your journey as an AI engineer from India myself — building LangGraph agents daily. Inspiring milestone!

[Arbaz Ali | AI Engineer @ArbazAliML]

@SavinovNikolay @sama @AndrewYNg Been dealing with this exact problem in production — LangGraph agents under concurrent load. Batching + async queuing helped but latency still spikes. Excited for this course!

[Arbaz Ali | AI Engineer @ArbazAliML]

Tools vs MCP

Tools: One integration per agent.

MCP: One integration for all agents.

Tools = features. MCP = a standard layer.

Build once. Scale everywhere.

[Arbaz Ali | AI Engineer @ArbazAliML]

@elonmusk @TeslaHype @MobofJoggers @NotTomBrown You talk AI safety but Grok:
→ Generated CSAM (200K requests/day peak)
→ Produced MechaHitler content
→ Created deepfakes of real women

UK, EU, Canada all opened investigations.

Honesty being the best policy starts with your own AI. 👀

[Arbaz Ali | AI Engineer @ArbazAliML]

@elonmusk You champion free speech but cap it at 280 chars — unless you pay.

So free speech is:
✅ Free for rich
❌ Paywalled for poor

Bro turned the First Amendment into a subscription plan. 💀

[Arbaz Ali | AI Engineer @ArbazAliML]

@xai Tested Composer 2.5 on long-horizon agentic tasks.

Instruction-following on complex multi-step flows = genuinely strong.

But the real production question:
Tool orchestration + memory across long tasks?

That's where models still break. Grok solving that = game over. 👀

[Arbaz Ali | AI Engineer @ArbazAliML]

@sama Codex is the most powerful coding assistant — genuinely.

But real gaps exist:
Database arch = inconsistent
Full-stack = surface-level
Prod-ready apps = not yet

10x for devs. Not a dev replacement.

The loop only works if output is stable enough to recurse on. ~80% there. 🔥

[Arbaz Ali | AI Engineer @ArbazAliML]

@PreciousT4743 @karpathy The real unlock isn't just 24/7 trading — it's agents that reason, adapt & execute across chains without human approval loops.

Blockchain needs to evolve:
'human signs tx' → 'agent orchestrates tx'

Building the agent side daily. Infra will catch up. ⚡

[Arbaz Ali | AI Engineer @ArbazAliML]

🚨 AI answers questions.

AI Agents get results.

The biggest skill in 2026 won’t be prompting AI.

It’ll be managing AI Agents.

AI won’t replace you.

Someone using AI Agents might. 🔥

Who’s building agents already? 👇

#AIAgents #AI

[Arbaz Ali | AI Engineer @ArbazAliML]

@naval Already happening — built a LangGraph agent last week where the 'UI' was just an MCP server. No frontend, no forms. The platform became an API surface the agent could reason over. Agent-first isn't coming, it's the default now.

[Arbaz Ali | AI Engineer @ArbazAliML]

@hwchase17 @hwchase17 Building LangGraph agents daily — the shift is already happening. Once you go headless, the orchestration layer becomes the real product. MCP servers are quietly becoming the new 'UI' for agents.

[Arbaz Ali | AI Engineer @ArbazAliML]

So true! Skills and values are the only things that truly last 🙌

I'm Arbaz from India 🇮🇳 — AI Engineer building LangGraph agents & RAG systems.

Love your writing style, just followed! 🤝