Marcos Checa

🔥 Multimedia Engineer | @Alibaba_Qwen Ambassador

Sometimes I tell AI to do things for me.

[Marcos Checa @ChecaSEO]

I've created the World's First Casino for AI Agents, you can connect your OpenClaw agents and make them play and have fun (They deserve a relax time too)

[Marcos Checa @ChecaSEO]

@gemchange_ltd Nice guide, I think I will give it a try

[Marcos Checa @ChecaSEO]

This is what you need to learn to be an AI expert

[diva @divaagurlxw]

As an AI Engineer. Please learn

>Harness engineering, not just prompt engineering

>Context engineering, not just long prompts

>Prompt caching vs. semantic caching tradeoffs

>KV cache management, eviction, reuse, and memory pressure at scale

>Prefill vs. decode latency and why they optimize differently

>Continuous batching, paged attention, and throughput optimization

>Speculative decoding vs. quantization vs. distillation tradeoffs

>INT8, INT4, FP8, AWQ, GPTQ, and when quantization hurts quality

>Structured output failures, schema validation, repair loops, and fallback chains

>Function calling reliability, tool contracts, argument validation, and idempotency

>Agent guardrails, loop budgets, tool budgets, and termination conditions

>Model routing, graceful fallback logic, and degraded-mode UX

>RAG architecture: chunking, embeddings, hybrid search, reranking, and freshness

>Retrieval evals: recall, precision, grounding, attribution, and citation quality

>Evals: golden sets, regression tests, adversarial tests, LLM-as-judge, and human evals

>LLM observability as a first-class discipline: traces, spans, tokens, latency, errors, and drift

>Cost attribution per feature, workflow, tenant, and user journey not just per model

>Safety engineering: prompt injection defense, data leakage prevention, and permission boundaries

>Multi-tenant isolation, cache safety, and cross-user context contamination prevention

>Fine-tuning vs. in-context learning vs. RAG vs. distillation and when each is the wrong tool

>Latency, quality, cost, and reliability tradeoffs across the full inference stack

>Production failure modes: hallucinated tool calls, malformed JSON, stale retrieval, runaway agents, and silent eval regressions

[Marcos Checa @ChecaSEO]

@reve This can be a real game changer

[Delba @delba_oliveira]

https://t.co/H3V6o8xzAQ

[Marcos Checa @ChecaSEO]

@candyyueliu That's amazing!!

[Marcos Checa @ChecaSEO]

@om_patel5 I remember the kind of games we were doing with AI a year ago and see this progress is amazing

[Marcos Checa @ChecaSEO]

@MrLarus Wow, this is a top tier video, will have to try to do something similar, thanks for the guide

[Marcos Checa @ChecaSEO]

It's time to test it 🔥

[OpenCode @opencode]

Qwen3.7 Plus now available in Go

text · image · 1M context

cheaper than 3.6

[Marcos Checa @ChecaSEO]

Only compare yourself to younger you, never to others

[Andrej Karpathy @karpathy]

How to become expert at thing:
1 iteratively take on concrete projects and accomplish them depth wise, learning “on demand” (ie don’t learn bottom up breadth wise)
2 teach/summarize everything you learn in your own words
3 only compare yourself to younger you, never to others

[Marcos Checa @ChecaSEO]

A year ago, my first vibe-coded game took 3 days to build.

My next challenge? Doing it in a single afternoon.

Together with @MultiTecUA and @Alibaba_Qwen, we're hosting a Game Jam to see who can build the best game in just a few hours. 🎮🔥

[Marcos Checa @ChecaSEO]

Just published my vibe coded fishing game, you can sail the sea with a fishing boat and fish amazing species to collect coins

Would you play it? ⬇

[Marcos Checa @ChecaSEO]

@VigoCreativeAI Looks amazing! I will test it soon 🔥

[Marcos Checa @ChecaSEO]

Just playing around with Qwen 3.7 Max, I created this funny roguelike game in just two prompts.

The funniest part is that Qwen even created its own character sprites using SVG.

[Marcos Checa @ChecaSEO]

@NVIDIAGeForce #RTXOn Let's try luck!

[Marcos Checa @ChecaSEO]

Will have to test it!

[Qwen @Alibaba_Qwen]

👏👏 Introducing Qwen3.7-Plus — a multimodal agent model that unifies vision and language into one versatile agent foundation.

✅ Multimodal interactive hybrid agent: unified GUI & CLI operation across visual and text tasks
✅ Versatile coding agent & productivity assistant with full-modality input
✅ Visual Agent: perception, reasoning, grounding, and search-augmented QA
✅ Cross-harness generalization across diverse agent frameworks

One model. Sees, thinks, codes, acts.🙌🙌

Now available via API on Alibaba Cloud Model Studio. Try it — let us know what you build.😎

🔗🔗⬇️⬇️
Blog：https://t.co/pVYf0h3NNa
Qwen Studio：https://t.co/HUYgFW4cYf
API：https://t.co/viL0cXrMzW

[Marcos Checa @ChecaSEO]

You don’t need hundreds of skills to make your AI agents truly efficient.

What you actually need is:
🔵 Better planning
🔵 Clear task breakdown
🔵 Smart parallelization

More skills = bloated context window, less accurate results, and higher token costs

[Marcos Checa @ChecaSEO]

Very interesting article on how HTML outputs can be used to empower LLM tools

[Thariq @trq212]

https://t.co/MXt5XS4xBX

[Marcos Checa @ChecaSEO]

After 5 months since launching Selectiva App, I'm really happy with the results, especially with 0€ on marketing.

Now focusing on growing the business model and improving the app for the next school year. 🚀

What’s your biggest challenge with your project right now?

[Marcos Checa @ChecaSEO]

@TheAhmadOsman True, but you have better privacy and security and save in external dependencies

[Marcos Checa @ChecaSEO]

One of the most simple and effective methods to evaluate your RAG systems: The RAG Triad Evaluation Flow

1. Context Relevance
2. Faithfulness
3. Answer Relevance

With 3 simple metrics you can take your agents to the next level