learning to engineer intelligence // 21 // Ex Intern @iitrpr, @IndianNavy
Delhi, India
Joined June 2024
- Tweets998
- Following2.6K
- Followers807
- Likes3.6K
Om retweeted
The latest Qwen Code (v0.26.3) already has the Qwen4 lineup baked into the code. Every model starting with qwen4 defaults to a 1M-token context.
Named in testing: qwen4-max, qwen4-plus, qwen4-coder-plus, plus dated ones like qwen4-max-2026-10-01, and even qwen4.5-max is prewired.
This is a preset in the official coding tool, not an official Apsara announcement. Amsterdam kicks off today; something should move this afternoon.
Feels close.
alibabacloud.com/apsara-conf…
Hoping the conference drops a surprise.
Then what's stopping me from earning this much?🤔
creators.unfair.so/worth/piu…
Here's a cure to your brain rot feed. Thank me later
historicaltechtree.com/
Great stuff, @adxtyahq
This is beautiful. Hope more of our learning resources get this 3B1B treatment.
I've been wanting to revise AI engineering concepts anyway. Seems like I've got the resource sorted out.
Today, I'm launching llmdojo.dev:
- The hands-on way to learn AI engineering: LLMs, RAG, agents, evals and inference.
- 50+ interactive labs. Tweak tokenizers, attention, embeddings and KV caches live.
- Train a neural network in your browser and watch it learn.
- Build agentic workflows with tool calls, memory, guardrails and bounded autonomy.
- Ship RAG that works: chunking, hybrid search, reranking, and LLM-as-a-judge evals.
- Production AI: model routing, fallbacks, quantization, caching and cost per request.
- 490+ interview questions, reported from OpenAI, Anthropic, Google, Meta, Amazon, Microsoft, Mistral and 50+ more.
- Every answer is cited from papers and books, or measured on real models.
