Circular Economy Lead @Stellantis 🚗 | Building the future of Mobility | TCG Player 🃏 & Dragon Ball Fan 🐉 | Personal Account
Lyon
Joined March 2012
- Tweets11.4K
- Following1.8K
- Followers814
- Likes17.5K
Pika 👺🇹🇭 retweeted
You need to know if people really want what you are building.
Lenny’s startup, Localmind, had users who said they loved it. But the problem it solved was small. It turned out to be a novelty.
His test for real demand is behaviour, not words. Look for people who:
- pay you
- feel strongly about the problem
- find you without you asking them
- keep using your product even when it is still rough
You can prepare user interviews in the same way.
In each interview, ask:
- what they have already tried
- what pushed them to start looking for a solution
- what kept them using the old way for so long
Score the call based on what they did, not what they say they might do.
Give your agent these 2 skills so they can do this with you:
- spot real demand from behaviour, not opinions
- run and score interviews based on what people have already done
Pika 👺🇹🇭 retweeted
La « star » LFIste du moment, qui a réussi hier à presque faire pleurer dans les chaumières via une vidéo, en expliquant que ses voyages de luxe c’est grâce à des amis riches, ceux sur lesquels LFI dégueule à longueurs de journées, et dont elle explique elle que ce n’est pas très grave qu’ils se fassent détruite leur outils de production par des manifestants.
Elle explique aussi que sa mère, qui vit avec 4 enfants dans le 15eme arrondissement de Paris, un des plus chers de la capitale, est vraiment obligée de s’acheter des gâteaux pour manger le midi, car elle est à découvert le 10/12 du mois.
Mais pourquoi donc ne pas faire comme les autres « pauvres », en allant vivre dans une HLM en grande banlieue, et en prenant chaque jour le RER ?
Quant à ses cours à 53€ les 50 minutes, il ne lui reste que 15/20€ en réalité, la marmotte lui prenant le reste.
Sur la première photo, un iPhone Pro Max, le modèle le plus cher, et en bandoulière un sac à 5500 balles, logique pour une femme de « gauche » en lutte contre le capitalisme et l’exploitation de l’homme par l’homme.
Pour les fringues et les bottes, à première vue elles ne viennent pas de chez la Foir’fouille… on peut supposer à première vue que ça dépasse les 20€.
Je pense qu’en fait ce qui serait bien dans son cas, c’est de se faire un peu oublier.
Pika 👺🇹🇭 retweeted
Google vient de lancer Playground, un site web qui permet à n'importe qui de créer des jeux vidéo en utilisant des prompts en langage naturel. presse-citron.net/creer-jeu-…
Pika 👺🇹🇭 retweeted
L'agence française de la transition écologique publie un avis qui explique pourquoi la voiture électrique a une empreinte carbone plus faible, lorsqu'elle est rechargée en France, même si sa fabrication produit plus d'émissions. presse-citron.net/voiture-el…
Pika 👺🇹🇭 retweeted
AI PM OS /skill of the day:
/opportunity (it's actually several skills in a sequence): run all your existing research through this process to produce a sequenced MECE roadmap of customer opportunities.
Not one-shotting so you get a chance to input, but it does a lot of the heavy lifting for you.
Pika 👺🇹🇭 retweeted
There's a big gap between solving customer problems and building an effective product strategy that many product leaders never learn to close.
When you're a junior PM solving customer problems an effective process is to:
Gather data → Identify opportunities → Prioritise and solve those opportunities.
This works well in the near term when you're trying to solve for today.
But it breaks down when you scale it up to the strategy level.
Because you've only identified half the picture.
What you've discovered is what is true TODAY, not what might happen TOMORROW.
And the second part is key the futher you project into the future - and it's a key reason why strategy is hard!
Because product strategy is all about placing bets on the future, you need to ALSO have a hypothesis on where you think the market, users, technology, the world, the company, etc are headed.
Which means you need to mix;
Data + Hypothesis = 'Strategic Narratives’.
Or as Roger L Martin, author of Playing to Win says;
"Strategy has a theory."
This hypothesis is the backbone of your product strategy - and it may go against what the data is saying today!
And that's the point.
Many successful strategies have been the result of betting against the trend.
Pika 👺🇹🇭 retweeted
Opus 5.5 designing LEGO 👀
I asked it to design a Microduck I can build with real LEGO pieces. It:
> designed it life-size using 1113 real LEGO parts
> verified: 3,204 connections, 0 collisions, every step buildable, centre of mass inside the feet 🤯
> made a 141-page LEGO-style booklet (237 steps)
> priced every piece in the browser and prepared the orders on BrickLink
Pika 👺🇹🇭 retweeted
What skills do you run in Claude/Codex when product decisions get messy?
Mine:
1. /mckinsey-issue-tree xitter.cf/nurijanian/status/2081…
2. /causal-tree xitter.cf/nurijanian/status/2082…
3. /first-principles
Drop yours, let's see what you got🕹️
Pika 👺🇹🇭 retweeted
A new WhatsApp Business MCP server lets developers use AI coding agents like Claude, Cursor, Codex, and ChatGPT to handle setup, messaging templates, testing, and troubleshooting.
spr.ly/6012BGrPgG
Pika 👺🇹🇭 retweeted
Anthropic just reliesed the best 1-hour workshop on Loop engineering: from one prompt to a Loop that improves itself
5:31 - loop memory
11:02 - check, build, commit
13:06 - one feature per loop
33:03 - MVP in 1 hour
Free, the best thing on Loop engineering I've come across.
Watch it, then try your first loop with the step-by-step guide below
Pika 👺🇹🇭 retweeted
Most AI agents don't have memory.
They have retrieval + prompt inflation.
RAG only looks information up. But a memory system decides what's worth keeping:
• policies
• preferences
• facts
• episodes
• traces
Then rebuilds the prompt from that every turn instead of stuffing more into it.
The missing piece isn't a bigger context window. It's a write path.
Bookmark this.
FREE course on how to build one in the comments ↓
🤝 Paid partnership
Pika 👺🇹🇭 retweeted
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
Pika 👺🇹🇭 retweeted
T-minus 30 minutes ⏰ until we're live!
Today we're talking about 'How AI Actually Helps in Discovery (and Where It Doesn't)'
I'll be sharing behind the scenes all the automations and everything I use AI for in discovery.
... and more importantly what I don't use it for.
Come hang out 👉 youtube.com/watch?v=Sh5T0bVB…
Pika 👺🇹🇭 retweeted
We’re excited to announce that Antonio Filosa, CEO of @Stellantis, will take the stage at #CES2027.
Filosa will showcase how the company, home to 14 brands worldwide, is making mobility smarter and more personal: ces.tech/press-releases/stel…
Pika 👺🇹🇭 retweeted
This guy literally explained why some people become successful while other stay average.
The reason is uncomfortable. Game Theory
Watch this:
Pika 👺🇹🇭 retweeted
Ex-Google Jeff Dean just released the best 1-hour lecture on AI engineering: from basics to Graphs
1:45 - LLM from scratch
17:22 - How to use AI models
30:03 - Prompt engineering
52:35 - One human coordinating 100 agents
27 years of building AI at Google, compressed into one hour.
Watch it, then read the full guide on Graph engineering below.
Pika 👺🇹🇭 retweeted
Claude doesn't come with a manual.
So I wrote a simple guide (like IKEA instructions):
Most people open Claude, type a question, get a bad answer & decide it's "just another chatbot."
The problem is nobody handed you the instructions. So I wrote one, like flat-pack furniture.
Here's the whole thing:
1. Get the entire manual (for free)
Go to how-to-ai.guide. Subscribe for free.
Open the received email in your inbox.
Go to the link, download "Ikea Claude manual"
Keep it open in a second tab while you follow along.
2. The one sentence to end every prompt with - it makes you better at Claude than 99% of people
3. Goals vs. tasks: why "write me an email" gets a worse result than telling Claude what you want
4. The Claude rule that stops your chats getting slow, dumb, and expensive
5. How to give Claude a memory, so every chat already knows who you are
6. The expensive model nobody warns you about and when to use it
7. Connecting your Gmail and calendar, and the skills you build once and never rebuild
No coding. No folders. No jargon.
If you can build a bookshelf, you can do this.
Keep it open in a second tab while you follow along:
→ how-to-ai.guide.
I removed every word that wasn't necessary and drew the rest.
If your team is still typing one-line questions into a Ferrari, send them this.
(If you share it, mention the newsletter. Be nice.)
