jportell (@jportell) header
jportell (@jportell) profile picture

Particularly interested in business strategy issues and on building a strong culture. Loves entrepreneurship, innovation, challenges and running marathons :)

Catalonia, Spain
Joined September 2008
Photo by jportell (@jportell)Photo by jportell (@jportell)Photo by jportell (@jportell)Photo by jportell (@jportell)
🚨 This is absolute GOLD. The @AnthropicAI engineer who literally wrote "Building Effective Agents" just dropped a 14-minute masterclass. saves you months of headaches trying to figure this out alone. bookmark for the weekend + read @Av1dlive's great guide below 👇
41
808
11
5,319
909,878
jportell retweeted
Instead of watching a 2-hour movie, watch this podcast with Claude’s CEO.
14
77
2
310
26,460
Claude's unwillingness to continue conversations after the context window is full is very frustrating. I am okay losing early context if I am working interactively on a project & making progress, I am not okay with suddenly being cut off and being forced to start a new chat.
400
116
42
2,668
227,808
REVIEWS
Use AI to get a leg up on your competitors I built a prompt for @perplexity_ai Comet that harvests your competitors reviews, groups pain points, and teaches you exactly where they're falling short so you can position your product better What you’ll uncover: - Repeating failures (fit, durability, shipping, UX) with source links - Frequency & severity of the issue plus its impact on revenue - Concrete fixes + claim lines you can own on PDP/ads All you have to do is copy-paste my template, fill in your competitors product info and click run Follow + Like + RT + comment “REVIEWS” and I’ll DM you the exact template. MUST BE FOLLOWING so I can DM you.
16
Thrilled to announce I’m starting pagemind.ai 🚀 We’re on a mission to make the web conversational, turning any site into an interactive, AI-driven experience that helps visitors understand, explore, and convert.
8
Multi-agent AI is a $50B lie. 99% of "multi-agent" systems are just single agents with fancy marketing. I just read the paper that exposes what real multi-agent intelligence actually looks like. Most people think multi-agent AI is just "multiple ChatGPTs in a room. That's like saying a surgical team is just "multiple people with knives." The real story is way deeper. Task allocation is completely broken. Current systems are basically throwing darts at a board. Give the math problem to whoever's free. Ask the creative agent to debug code. It's chaos disguised as intelligence. Real multi-agent systems need dynamic specialization. Not just "Agent 1 does X, Agent 2 does Y" but context-aware matching based on capability, workload, and past performance. The memory problem is insane. Single agents just track conversations. Multi-agent systems need five different memory types: short-term task state, long-term expertise, episodic collaboration history, consensus knowledge, and hierarchical access control. Most current systems give every agent amnesia between tasks. Context management is where everything breaks. Each agent needs to track three layers simultaneously: the big picture mission, their specific piece, and what everyone else is doing. Fail at any layer and the whole system becomes expensive nonsense. Game theory matters more than code. When agents debate or negotiate, you're not optimizing for "correctness." You're finding equilibrium states. The research shows Stackelberg dynamics work better than Nash equilibrium for most real tasks. Nobody talks about this because it's not as sexy as "look, the robots are talking." The applications they outline are wild. Agents that negotiate smart contracts autonomously. Fraud detection where different specialists hunt different attack patterns. Consensus mechanisms that actually think through decisions. We're not building better chatbots. We're building the foundation for autonomous economic systems. The gap between current "multi-agent" demos and actual multi-agent intelligence is massive. Real systems will have specialized roles, shared memory architectures, and game-theoretic coordination. They'll solve problems no individual agent can handle. Same principle that makes human teams work. Just faster, and at scale. Most of what people call "multi-agent" today is just single agents with fancy prompting. The companies that figure out real multi-agent coordination first will have a 10x advantage. Everyone else is building expensive theater.
64
194
22
1,139
121,317
Ens sentim tant estimats…
Esto le hacen a Vinicius y España declara 2 días de luto nacional. Pero como son a los jugadores del Barça B, no pasa nada.
1
3
165
Venga, voy a hablar sobre las elecciones catalanas. Cosa que odio (en serio, la política me parece aburrida y divisiva). Amigos del resto de España/Mundo: no lean esto. Mañana seguimos hablando de cosas interesantes. Empiezo: por qué NO hay que votar al PSC?
191
1,316
195
2,023
513,499
Plaça Lesseps fa més de 150 anys. Gràcia 1870. 📷 Troballa grossa aquesta a Alemanya, Hessisches Staatsarchiv Darmstadt.
118
1,153
52
5,157
345,904
“When we manage by outputs, we are asking our teams to build specific features, implement specific projects, or deliver on specific initiatives. We are dictating the actions of our teams.” - @ttorres bit.ly/31LSptM #prodmgmt #ux
1
40
No pharmeceutical can rival exercise when it comes to slowing the aging process - particularly the aging brain. A large-scale trial investiated th effect of aerobic exercise (running, cycling) on cognition in the 20- to 67-year age range. Here’s what they found: Better thinking skills as you age: Exercise helped people’s executive function - that’s your brain’s ability to manage tasks, plan, and focus. Thicker brain cortex, regardless of age: • Higher cognitive reserve- aiding task flexability and improvisation • Resilience against brain aging and neurodegenerative disease • Improved neural plasticity, essential for lifelong learning and skill acquisition • Bolsters emotional regulation Real-world Application: The way the workouts were implemented could be a practical recommendation for improving brain health for the general public. Consistency is key: participants exercised about 40 minutes, four times a week for six months. Gradually increasing effort: they started easy (55-65% max heart rate) and slowly increased their workout intensity (65%-75% max heart rate). Monitoring progress: using heart rate monitors helped them stay in the desired exercise intensity. Flexibility: participants could choose the exercise (running or cycling) they preferred and set their schedules. #everydayoctober
67
459
40
2,782
521,875
🔴 Seguimos horrorizados y exigimos el fin inmediato del derramamiento indiscriminado de sangre. "La violencia que estamos viendo no tiene precedentes. #Gaza está siendo arrasada y miles de personas están muriendo". Meinie Nicolai, directora general de #MSF 👇
188
5,550
230
7,163
522,444
The distributed idea suppression complex (DISC) is a concept from @ericweinstein that describes the phenomenon of widespread idea suppression. @elonmusk owning @Twitter shows a wrench at DISC, which is why the usual suspects are panicking. Good.
1
2
2
Por una vez que Twitter actúa con cierta urgencia y rapidez. Se ve que hay mucha pasta e intereses en el negocio de la cancelación de voces discrepantes.
2
1
3
Avui al blog us parlo d’aquest quadre d la Mare d Déu del #Vinyet d #Sitges obra d’Agustí Ferrer Pino, i de pas us faig una proposta… qui s’hi apunta? criticartt.blogspot.com/2021… @AjSitges @CulturaSitges @lecodesitges @radiomaricel @PlatgesSitges @TurismeDeSitges @MuseusDeSitges
8
1
18
Em passes l'oli?
37
4
2
89