..when I watch twitters steady flow of chaotic, short, repetitious arguments from around the world, it seems about as informative as modem binary transmission lights. Data is freely flowing, but soulless and without consciousness. Information, with severely limited comprehension.
@glassiniva@JerichoGTA True, but at the same time cartoonishly outrageous.
We’re nearly at Holodeck levels of realism, simulating a shitty mirror of our real world.
Am I the only one who’s not worried about the survival of the species?
I worry about friends, family and society. I worry about societies decay and corruption. Inhabiting more planets won’t save us from communism, trans humanism or Islamic conquest.
If we have other planets with humans on them, that doesn’t stop one planet and billions from perishing.
I’m not saying ambitions to head to the stars aren’t worth doing - but pinning it primarily on the survival of the species is misplaced.
If I had 6 months to become a GenAI Engineer, I wouldn't waste time collecting certificates or watching endless tutorials. I'd follow a roadmap that turns knowledge into working systems.
First, master the foundations. Learn Python, FastAPI, asyncio, Pydantic, multimodal LLMs, context engineering, structured outputs and advanced RAG. Understand how models process information before trying to build complex agents.
Next, learn to build agents that survive the real world. Master MCP, multi-agent orchestration, durable workflows, human approval gates, GraphRAG and generative UI. Then add evaluation, automated regression tests, observability and AI safety. A working demo is easy. A reliable agent is engineering.
The next level is infrastructure. Learn vLLM, SGLang, KV-cache optimization, speculative decoding, quantization, semantic caching and cost control. Track latency, token usage and failures. If you cannot measure performance or control costs, you cannot build production-grade AI.
Finally, ship open-source projects, publish architecture breakdowns, document real engineering failures, and apply for GenAI Engineer, Agent Engineer or AI Infrastructure roles. Six months of building can teach you more than years of passively watching tutorials. Most people consume AI knowledge. Builders turn it into careers.