nicco

writing about ai & crypto

finding signal in the noise // dm open

[nicco @0xnicc0]

an Anthropic engineer just said something wild about prompting: "at Anthropic, we don't write prompts anymore, we build loops"

in 42 minutes, she shows how the Claude team creates loops that prompt themselves

if this cost $400, people would call it the best agents course of the year

it's free

watch it, then use the step-by-step guide below to build your first loop

[nicco @0xnicc0]

Andrew Ng just dropped a 1-hour course that builds agentic knowledge graphs from scratch

09:11 - Query connections with Cypher
12:28 - Model your data as a knowledge graph
24:09 - Build your first agent with Google ADK
45:58 - Delegate work to sub-agents
58:15 - Give your agents memory with session state

worth more than 10 articles on loop engineering

save this and watch it tonight

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

this massive resource drop from Google's former CEO says the quiet part out loud

building wealth is becoming easier if you actually learn how to use the tools - not by scrolling AI threads, but by understanding agents, Claude Code, prompts, memory, skills, MCP, and routines

everything below is free:

- agent architecture
- Claude Code 101
- Claude Code in action
- prompt engineering
- interactive prompt course
- Claude.md + memory
- MCP
- Claude Code ultimate guide
- awesome Claude Code
- Anthropic academy
- official Claude Code docs

all of this costs $0

most people will keep asking AI one question at a time

[nicco @0xnicc0]

an Anthropic engineer just said something wild about prompting: "at Anthropic, we don't write prompts anymore, we build loops"

in 42 minutes, she shows how the Claude team creates loops that prompt themselves

if this cost $400, people would call it the best agents course of the year

it's free

watch it, then use the step-by-step guide below to build your first loop

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

this Anthropic Research Lead's take on agentic graphs is wild

"99% of our engineers run swarms of 300+ self-improving agents, now everyone is building agentic graphs"

in a 20-minute session, an Anthropic team member breaks down how graph engineering turns isolated agents into systems that improve themselves

the real setup is Claude running through graph workflows, plan mode, and dynamic orchestration

agents → loops → graphs → self-improving systems

better than most $300 graph engineering courses

save this and watch the talk, then read the article below

[nicco @0xnicc0]

this massive resource drop from Google's former CEO says the quiet part out loud

building wealth is becoming easier if you actually learn how to use the tools - not by scrolling AI threads, but by understanding agents, Claude Code, prompts, memory, skills, MCP, and routines

everything below is free:

- agent architecture
- Claude Code 101
- Claude Code in action
- prompt engineering
- interactive prompt course
- Claude.md + memory
- MCP
- Claude Code ultimate guide
- awesome Claude Code
- Anthropic academy
- official Claude Code docs

all of this costs $0

most people will keep asking AI one question at a time

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

this Anthropic engineer's shift to agentic GRAPHS is nuts

"90% of our engineers were already running self-improving loops, now everyone is building agentic graphs"

"no more prompting"

in just 10 minutes, she builds her complete Claude Code system live from an empty terminal

agents → loops → graphs → self-improving systems

this 10-minute video is worth more than most $1,000 agent engineering courses

save the full guide below before everyone catches up

[nicco @0xnicc0]

this Anthropic Research Lead's take on agentic graphs is wild

"99% of our engineers run swarms of 300+ self-improving agents, now everyone is building agentic graphs"

in a 20-minute session, an Anthropic team member breaks down how graph engineering turns isolated agents into systems that improve themselves

the real setup is Claude running through graph workflows, plan mode, and dynamic orchestration

agents → loops → graphs → self-improving systems

better than most $300 graph engineering courses

save this and watch the talk, then read the article below

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

Andrej Karpathy spent 8 years at OpenAI and Tesla

last week, he put everything he knows into one free 2-hour lecture

people pay $15k for bootcamps that teach half of this

you probably don't have 2 hours right now

save this and watch it tonight, then build your first LOOP

[nicco @0xnicc0]

this Anthropic engineer's shift to agentic GRAPHS is nuts

"90% of our engineers were already running self-improving loops, now everyone is building agentic graphs"

"no more prompting"

in just 10 minutes, she builds her complete Claude Code system live from an empty terminal

agents → loops → graphs → self-improving systems

this 10-minute video is worth more than most $1,000 agent engineering courses

save the full guide below before everyone catches up

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

my friend applied to 200 tech jobs in two years

no CS degree. not one callback

last month, Anthropic offered him $750,000

all because of one free Stanford lecture

one hour. a Stanford professor explains how ChatGPT actually works. not the Twitter version, the real one

he watched it in bed, paused it eleven times

after that hour, he told me something I didn't believe: "it's embarrassingly simple"

three days later, he applied to Anthropic. every single question they asked him, he knew from that video

save this and watch it tonight

[nicco @0xnicc0]

Andrej Karpathy spent 8 years at OpenAI and Tesla

last week, he put everything he knows into one free 2-hour lecture

people pay $15k for bootcamps that teach half of this

you probably don't have 2 hours right now

save this and watch it tonight, then build your first LOOP

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

OpenAI CEO Sam Altman: "you no longer need to write prompts yourself"

in less than 35 minutes, he explains how to use ChatGPT better than most people

he originally delivered this lecture to Stanford students

my friend sent me the recording

I watched it last night and realized I was using less than 15% of what ChatGPT can actually do

watch it, then read the guide below to build a system that prompts itself

[nicco @0xnicc0]

my friend applied to 200 tech jobs in two years

no CS degree. not one callback

last month, Anthropic offered him $750,000

all because of one free Stanford lecture

one hour. a Stanford professor explains how ChatGPT actually works. not the Twitter version, the real one

he watched it in bed, paused it eleven times

after that hour, he told me something I didn't believe: "it's embarrassingly simple"

three days later, he applied to Anthropic. every single question they asked him, he knew from that video

save this and watch it tonight

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

my friend makes $1.2 million a year as an Anthropic engineer

I asked him how he got so good at prompting

he sent me a video that was never supposed to get out

their core team's actual prompting playbook

I watched it last night

halfway through, I realized I've been using Claude completely wrong for two years

watch it and save it

[nicco @0xnicc0]

OpenAI CEO Sam Altman: "you no longer need to write prompts yourself"

in less than 35 minutes, he explains how to use ChatGPT better than most people

he originally delivered this lecture to Stanford students

my friend sent me the recording

I watched it last night and realized I was using less than 15% of what ChatGPT can actually do

watch it, then read the guide below to build a system that prompts itself

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

this one is WILD - Anthropic leaked a 6 agent GRAPH that turns one prompt into a finished PR

your name appears once, at the end, for five minutes

spec → planner → three builders → critic → scribe → PR → back into the spec

the planner gets one shot. every agent downstream inherits that decision

a bad plan executed by five perfect agents is still a perfect failure

this is the layer most teams keep inside one giant prompt

three builders run in PARALLEL because the planner cuts work at the dependency boundary

if one lane needs another lane to finish first, the graph was wrong before the coding started

tests own the local loop. red goes back to the builder, green unlocks the critic

the critic never reviews a broken build

tests ask can it run. the critic asks should it exist

the scribe writes the PR from the trace, not from memory

remove that seat and every PR becomes the same vague summary nobody trusts

the back edge is the entire trick: today's failures become tomorrow's constraints without another human prompt

one human step remains: approve or send it back

five minutes instead of five hours. one prompt in, one finished PR out

[nicco @0xnicc0]

my friend makes $1.2 million a year as an Anthropic engineer

I asked him how he got so good at prompting

he sent me a video that was never supposed to get out

their core team's actual prompting playbook

I watched it last night

halfway through, I realized I've been using Claude completely wrong for two years

watch it and save it

[seeco @seeconvm]

https://t.co/0B93pIbwuQ

[nicco @0xnicc0]

your agent is not a loop

it is a graph that can touch production

a loop gives you repetition. production gives you branches, checkpoints, interrupts, tools, retries and side effects

that is not a loop anymore. that is a distributed state machine with an LLM inside

THE GRAPH ENGINEERING STACK

01 - topology is code
every node needs an input contract, an output contract and a legal next state
if the route exists only inside the model, it does not exist

02 - state must survive the process
memory in a prompt is context, memory in a graph is a checkpoint you can replay, inspect and recover
if one worker crash erases the run, you built a demo

03 - every run needs a map
the final answer cannot tell you which branch wasted the budget
trace nodes, transitions, tools, latency, tokens and failures as one execution graph

04 - effects need boundaries
a legal edge can still call the wrong tool with the wrong permission
run generated execution somewhere built to be destroyed

05 - test the path not just the answer
the same output can come from a safe route or a catastrophic one
grade state transitions, tool choices, retries and exits

prompt engineering asks what should the model say. graph engineering asks what can happen next

that second question is where production starts

save this before drawing another circle with an arrow

[nicco @0xnicc0]

this one is WILD - Anthropic leaked a 6 agent GRAPH that turns one prompt into a finished PR

your name appears once, at the end, for five minutes

spec → planner → three builders → critic → scribe → PR → back into the spec

the planner gets one shot. every agent downstream inherits that decision

a bad plan executed by five perfect agents is still a perfect failure

this is the layer most teams keep inside one giant prompt

three builders run in PARALLEL because the planner cuts work at the dependency boundary

if one lane needs another lane to finish first, the graph was wrong before the coding started

tests own the local loop. red goes back to the builder, green unlocks the critic

the critic never reviews a broken build

tests ask can it run. the critic asks should it exist

the scribe writes the PR from the trace, not from memory

remove that seat and every PR becomes the same vague summary nobody trusts

the back edge is the entire trick: today's failures become tomorrow's constraints without another human prompt

one human step remains: approve or send it back

five minutes instead of five hours. one prompt in, one finished PR out

[seeco @seeconvm]

https://t.co/0B93pIbwuQ