seeco

@seeconvm • 3,626 subscribers

ai researcher post about AI / crypto open for collabs

Shorts

Opus 5.5 just turned a simple idea into this I'm still in shock… Drop the PDF below into Opus 5.5 and try it yourself. Then read the article below, it's everything I learned with Opus 5.5.

Opus 5.5 just turned a simple idea into this I'm still in shock… Drop the PDF below into Opus 5.5 and try it yourself. Then read the article below, it's everything I learned with Opus 5.5.

147,326 views

Opus 5.5 is a motion designer now It’s f*cking unreal… Give the PDF below to Opus 5.5 and watch it work. Then read the guide below, it’s the easiest way to get started.

Opus 5.5 is a motion designer now It’s f*cking unreal… Give the PDF below to Opus 5.5 and watch it work. Then read the guide below, it’s the easiest way to get started.

166,466 views

One prompt in Opus 5.5, zero After Effects It’s f*cking unreal… Paste the PDF below into Opus 5.5 and see what you get. Then save the guide below, it’s all you need to make videos like this.

One prompt in Opus 5.5, zero After Effects It’s f*cking unreal… Paste the PDF below into Opus 5.5 and see what you get. Then save the guide below, it’s all you need to make videos like this.

59,452 views

Opus 5.5 did this in one try, not After Effects I’m still shocked… Send the PDF below to Opus 5.5 and copy what I did. Then read the guide below, it’s the shortcut I wish I had.

Opus 5.5 did this in one try, not After Effects I’m still shocked… Send the PDF below to Opus 5.5 and copy what I did. Then read the guide below, it’s the shortcut I wish I had.

27,861 views

loop engineering worked for 6 months. then one pipeline forced a graph a loop is one agent on discover → plan → execute → verify, until a stop condition hits enough for most agent work: failing tests, draft + self-review, inbox triage, one research writeup the break is nameable: if the thing that produced the output also grades it, you don't have quality control, you have a rubber stamp his numbers before the split: 40 drafts 31 passed self-review 18 actually publish-ready on manual read gap = 13 drafts with a passing grade from a biased judge he split one loop into 3 nodes - researcher · sources + notes only - writer · notes + task only · never grades itself - reviewer · separate model call · check the bar · write nothing same 40-draft workload. manual approval 18/40 → 33/40. reviewer rejected 14 with quotable reasons. 2 escalated to human instead of burning tokens forever approval 45% → 82.5%. inference cost +60% worth it when a bad draft costs more than three model calls. not worth it for "fix the failing test" playbook: - 1 law · producer never grades - 1 router · ship / retry / escalate - 1 state object · one owner per field · guarded writes - split the verifier first · cheapest highest-leverage move loop engineering → graph engineering. not five nodes for vibes, one split verifier when the loop starts grading its own homework full contracts, router, and decision function in the article

loop engineering worked for 6 months. then one pipeline forced a graph a loop is one agent on discover → plan → execute → verify, until a stop condition hits enough for most agent work: failing tests, draft + self-review, inbox triage, one research writeup the break is nameable: if the thing that produced the output also grades it, you don't have quality control, you have a rubber stamp his numbers before the split: 40 drafts 31 passed self-review 18 actually publish-ready on manual read gap = 13 drafts with a passing grade from a biased judge he split one loop into 3 nodes - researcher · sources + notes only - writer · notes + task only · never grades itself - reviewer · separate model call · check the bar · write nothing same 40-draft workload. manual approval 18/40 → 33/40. reviewer rejected 14 with quotable reasons. 2 escalated to human instead of burning tokens forever approval 45% → 82.5%. inference cost +60% worth it when a bad draft costs more than three model calls. not worth it for "fix the failing test" playbook: - 1 law · producer never grades - 1 router · ship / retry / escalate - 1 state object · one owner per field · guarded writes - split the verifier first · cheapest highest-leverage move loop engineering → graph engineering. not five nodes for vibes, one split verifier when the loop starts grading its own homework full contracts, router, and decision function in the article

14,534 views

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Fable 5.5 just f*cked the whole mecha anime industry... this is f*cking crazy i made a giant robot catch a meteor with Fable 5.5, no image generation, no video model, 100% code > 6 agent teams working in parallel > an AI art director on every team > 36 shots, 1,968 code-rendered frames > zero image generation, zero video models > every frame coded, not drawn full workflow + code in the comments then, read the complete motion design guide

Fable 5.5 just fcked the whole mecha anime industry... this is fcking crazy i made a giant robot catch a meteor with Fable 5.5, no image generation, no video model, 100% code > 6 agent teams working in parallel > an AI art director on every team > 36 shots, 1,968 code-rendered frames > zero image generation, zero video models > every frame coded, not drawn full workflow + code in the comments then, read the complete motion design guide

53,075 views • 5 days ago

A guy dumped 11 years of notes into one system overnight. 47,000 notes. 0 folders. By morning, every idea had found its neighbors. Old journal entries connected to recent projects. Forgotten research started appearing around current interests. The vault stopped acting like storage. It started acting like memory. The interesting part isn't the visualization. It's realizing you already have years of knowledge sitting in files you never revisit. Most people save information. A small group is building systems that surface the right idea at the right moment. Your second brain becomes useful when old thoughts start finding you again. Bookmark this before knowledge graphs become normal.

A guy dumped 11 years of notes into one system overnight. 47,000 notes. 0 folders. By morning, every idea had found its neighbors. Old journal entries connected to recent projects. Forgotten research started appearing around current interests. The vault stopped acting like storage. It started acting like memory. The interesting part isn't the visualization. It's realizing you already have years of knowledge sitting in files you never revisit. Most people save information. A small group is building systems that surface the right idea at the right moment. Your second brain becomes useful when old thoughts start finding you again. Bookmark this before knowledge graphs become normal.

34,953 views • 3 months ago

A systems architect fed Claude 9 years of notes. 4,102 files. Nearly a decade of memory. What came back wasn't a folder. It was a knowledge graph. Daily logs linked to old projects. Research connected to decisions made years later. Ideas that seemed forgotten suddenly had context again. The interesting part isn't storing information. It's seeing how everything connects. Most people have thousands of notes they never revisit. A small group is turning their notes into living systems that compound over time. A second brain stops being storage. It becomes memory you can actually navigate. Bookmark this before knowledge graphs become the default way to manage information.

A systems architect fed Claude 9 years of notes. 4,102 files. Nearly a decade of memory. What came back wasn't a folder. It was a knowledge graph. Daily logs linked to old projects. Research connected to decisions made years later. Ideas that seemed forgotten suddenly had context again. The interesting part isn't storing information. It's seeing how everything connects. Most people have thousands of notes they never revisit. A small group is turning their notes into living systems that compound over time. A second brain stops being storage. It becomes memory you can actually navigate. Bookmark this before knowledge graphs become the default way to manage information.

25,679 views • 3 months ago

A former Deliveroo courier built an AI server in a Nike shoebox $180 in used parts $8.50/month in electricity It replaced a $412/month AI stack Now it runs automations for multiple clients Most people think you need a startup to sell AI He built one from spare parts on a kitchen table Bookmark this before local AI becomes the cheaper way to work

A former Deliveroo courier built an AI server in a Nike shoebox $180 in used parts $8.50/month in electricity It replaced a $412/month AI stack Now it runs automations for multiple clients Most people think you need a startup to sell AI He built one from spare parts on a kitchen table Bookmark this before local AI becomes the cheaper way to work

21,189 views • 4 months ago

ONE OF CLAUDE'S DEVELOPERS TOLD ME: "YOU'RE USING AI AT 5%." I expected some crazy setup. It was just a folder with text files. Then he opened the graph. Thousands of notes. Everything connected. "I didn't link any of this. Claude did." Then he typed: where am i contradicting myself? 10 seconds later it returned old notes, dates, and quoted his own words back to him. He laughed: "No friend will ever tell you this. But this thing will." That changed how I think about AI. Most people use it like a chatbot. A few use it like a mirror.

ONE OF CLAUDE'S DEVELOPERS TOLD ME: "YOU'RE USING AI AT 5%." I expected some crazy setup. It was just a folder with text files. Then he opened the graph. Thousands of notes. Everything connected. "I didn't link any of this. Claude did." Then he typed: where am i contradicting myself? 10 seconds later it returned old notes, dates, and quoted his own words back to him. He laughed: "No friend will ever tell you this. But this thing will." That changed how I think about AI. Most people use it like a chatbot. A few use it like a mirror.

14,428 views • 3 months ago

No more content to load