Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Introducing RVM: The Virtual Machine Reimagined for the Agentic Age. Traditional virtual machines were designed for stable, predictable workloads. You carve up hardware into fixed slices, assign memory and CPU, and hope the boundaries hold. That model assumes applications sit still. Agents don’t. They spin up, disappear, coordinate, and...

rUv's profile picture

rUv

54,043 subscribers

14,334 görüntüleme • 5 ay önce •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

500 agents shouldn't be running all day. they should spin up for the exact window a task needs, finish it, and disappear until the next trigger. the bigger the swarm gets, the less any single agent matters. what matters is the system wiring them together. 100 agents can form up to 4,950 possible pairwise connections. 500 = 124,750. when a real signal lands, it spins up the whole workforce: 1 trigger → 100–500 Kimi agents → 5 live data feeds → up to 4,000 steps → verified artifact and that's before you add: sources → claims → memory → tool calls → contradictions → retries so the architecture has to fold all of that activity into one shared state, continuously. more agents just buys you more raw compute. the graph is the only thing standing between 124,750 possible connections and pure noise.
0:20

500 agents shouldn't be running all day. they should spin up for the exact window a task needs, finish it, and disappear until the next trigger. the bigger the swarm gets, the less any single agent matters. what matters is the system wiring them together. 100 agents can form up to 4,950 possible pairwise connections. 500 = 124,750. when a real signal lands, it spins up the whole workforce: 1 trigger → 100–500 Kimi agents → 5 live data feeds → up to 4,000 steps → verified artifact and that's before you add: sources → claims → memory → tool calls → contradictions → retries so the architecture has to fold all of that activity into one shared state, continuously. more agents just buys you more raw compute. the graph is the only thing standing between 124,750 possible connections and pure noise.

kocer

40,907 görüntüleme • 5 gün önce

FIVE LAYERS OF AGENT ENGINEERING, EACH ONE WRAPS THE ONE BELOW IT. IF YOU SKIP LAYER 2, YOUR LAYER 5 WILL LOOK BROKEN WHEN IT IS ACTUALLY JUST STANDING ON NOTHING. for weeks i debated harness vs loop vs graph like they were competing choices. then a stack diagram made the shape obvious. they are not choices. they are floors. 01 | prompt engineering. the message. unit of work: one input. inputs are role, instructions, examples, format. output is a single raw response. 02 | context engineering. the memory. unit of work: what stays in the window. a curator selects, compresses, and drops from query, docs, memory, prior turns, and tool outputs before the prompt runs. 03 | harness engineering. the machine. unit of work: the machine itself. gather (context + prompt) → LLM → tools or sub-agents → verifier → final response. the article calls this the operating environment. 04 | loop engineering. the system. unit of work: the run. goal + success criteria + max iterations + budget + completion check wrap around one harness pass. failed pass appends results to context and retries. 05 | graph engineering. the topology. unit of work: the graph run. goal + nodes + edges + state schema. graph routes to agent nodes, tool nodes, or human approval. a reviewer node with a different model and fresh context checks the final answer. the wrapping is the whole point. layer 5 assumes layer 4 works. layer 4 assumes layer 3 works. skip layer 2 and layer 3's verifier keeps failing without a clear reason. this is why swapping the model is a one-day project and swapping the stack is a quarter. the model is the commodity. the five layers around it are the engineering. full three-layer breakdown of the top of the stack (harness, loop, graph) in the post below.
0:20

FIVE LAYERS OF AGENT ENGINEERING, EACH ONE WRAPS THE ONE BELOW IT. IF YOU SKIP LAYER 2, YOUR LAYER 5 WILL LOOK BROKEN WHEN IT IS ACTUALLY JUST STANDING ON NOTHING. for weeks i debated harness vs loop vs graph like they were competing choices. then a stack diagram made the shape obvious. they are not choices. they are floors. 01 | prompt engineering. the message. unit of work: one input. inputs are role, instructions, examples, format. output is a single raw response. 02 | context engineering. the memory. unit of work: what stays in the window. a curator selects, compresses, and drops from query, docs, memory, prior turns, and tool outputs before the prompt runs. 03 | harness engineering. the machine. unit of work: the machine itself. gather (context + prompt) → LLM → tools or sub-agents → verifier → final response. the article calls this the operating environment. 04 | loop engineering. the system. unit of work: the run. goal + success criteria + max iterations + budget + completion check wrap around one harness pass. failed pass appends results to context and retries. 05 | graph engineering. the topology. unit of work: the graph run. goal + nodes + edges + state schema. graph routes to agent nodes, tool nodes, or human approval. a reviewer node with a different model and fresh context checks the final answer. the wrapping is the whole point. layer 5 assumes layer 4 works. layer 4 assumes layer 3 works. skip layer 2 and layer 3's verifier keeps failing without a clear reason. this is why swapping the model is a one-day project and swapping the stack is a quarter. the model is the commodity. the five layers around it are the engineering. full three-layer breakdown of the top of the stack (harness, loop, graph) in the post below.

kocer

30,675 görüntüleme • 17 gün önce

The two authors of Pi share three design principles in this 3-minute video: → Coding doesn’t need a separate memory system Code is the ground truth, and it keeps changing. Maintaining a separate memory that describes the code creates one more thing to maintain and keep in sync. Models can usually infer a project’s structure and coding style after reading one or two files. If they need help navigating it, a folder list with short descriptions is enough. For a chat bot, the ground truth is the conversation log. They append questions and answers from Slack channels to JSONL, then query the history with jq. → Bash is all you need Models already know Bash and can combine commands and scripts themselves. They don’t use MCP for Sentry data. Instead, they built an on-demand Skill that handles authentication and pulls the data in a fixed format. → Load tool output into context only when needed When Sentry returns 52 items, the Skill puts only 3 into context and writes the rest to a JSON file. Once the model confirms the structure looks right, it reads the remaining items.
2:56

The two authors of Pi share three design principles in this 3-minute video: → Coding doesn’t need a separate memory system Code is the ground truth, and it keeps changing. Maintaining a separate memory that describes the code creates one more thing to maintain and keep in sync. Models can usually infer a project’s structure and coding style after reading one or two files. If they need help navigating it, a folder list with short descriptions is enough. For a chat bot, the ground truth is the conversation log. They append questions and answers from Slack channels to JSONL, then query the history with jq. → Bash is all you need Models already know Bash and can combine commands and scripts themselves. They don’t use MCP for Sentry data. Instead, they built an on-demand Skill that handles authentication and pulls the data in a fixed format. → Load tool output into context only when needed When Sentry returns 52 items, the Skill puts only 3 into context and writes the rest to a JSON file. Once the model confirms the structure looks right, it reads the remaining items.

Vox

166,853 görüntüleme • 29 gün önce

STANFORD BUILT AN LLM MEMORY THAT WORKS LIKE A BRAIN AND IT MADE STANDARD RAG LOOK ANCIENT no fine-tuning. no bigger model. just a smarter way to remember the insight: the brain doesn't store memories as isolated notes. it indexes them and links them - so one cue pulls back the whole memory so HippoRAG stops treating documents as separate chunks. it builds a knowledge graph, then spreads activation across it (Personalized PageRank) to pull a full answer from a single cue - the brain calls this "pattern completion" the result: > up to 20% better on hard multi-hop questions > 10-30x cheaper and 6-13x faster than iterative retrieval > connects facts split across documents in a single step most RAG still encodes every passage in isolation - so it breaks the moment the answer is scattered more retrieval was never the answer. brain-like memory is bookmark this. the next leap in AI agents is how they remember, not how big they get
0:29

STANFORD BUILT AN LLM MEMORY THAT WORKS LIKE A BRAIN AND IT MADE STANDARD RAG LOOK ANCIENT no fine-tuning. no bigger model. just a smarter way to remember the insight: the brain doesn't store memories as isolated notes. it indexes them and links them - so one cue pulls back the whole memory so HippoRAG stops treating documents as separate chunks. it builds a knowledge graph, then spreads activation across it (Personalized PageRank) to pull a full answer from a single cue - the brain calls this "pattern completion" the result: > up to 20% better on hard multi-hop questions > 10-30x cheaper and 6-13x faster than iterative retrieval > connects facts split across documents in a single step most RAG still encodes every passage in isolation - so it breaks the moment the answer is scattered more retrieval was never the answer. brain-like memory is bookmark this. the next leap in AI agents is how they remember, not how big they get

NO1ennn

32,454 görüntüleme • 1 ay önce

🚨ICE/Border Patrol agents were caught on camera committing a hit-and-run in Minneapolis. In the video, an unmarked federal vehicle is stopped in the road. A legal observer is sitting in a car behind them, not violating the law, just observing. Then, without warning, the agents start reversing the vehicle towards the observer… But, they failed to notice a parked car sits directly behind them. As the observer backs up, the federal vehicle slams into the parked car. They stop… and pull forward. You can see the moment they realized they just fucked up, and are debating what to do. They realize no one is in it… so they drive away. And this isn’t the first time… ICE and Border Patrol agents are driving aggressively through communities across the country, using their vehicles as weapons, putting neighbors, families, and bystanders at risk. And anytime they hit someone, instead of a parked car, they aggressively arrest the person they hit and say they were “rammed.” For a department that constantly claims it’s hunting criminals, their agents sure seem to be the ones committing the crimes.
0:36

🚨ICE/Border Patrol agents were caught on camera committing a hit-and-run in Minneapolis. In the video, an unmarked federal vehicle is stopped in the road. A legal observer is sitting in a car behind them, not violating the law, just observing. Then, without warning, the agents start reversing the vehicle towards the observer… But, they failed to notice a parked car sits directly behind them. As the observer backs up, the federal vehicle slams into the parked car. They stop… and pull forward. You can see the moment they realized they just fucked up, and are debating what to do. They realize no one is in it… so they drive away. And this isn’t the first time… ICE and Border Patrol agents are driving aggressively through communities across the country, using their vehicles as weapons, putting neighbors, families, and bystanders at risk. And anytime they hit someone, instead of a parked car, they aggressively arrest the person they hit and say they were “rammed.” For a department that constantly claims it’s hunting criminals, their agents sure seem to be the ones committing the crimes.

Jesus Freakin Congress

814,594 görüntüleme • 7 ay önce

A very prominent African shared this video with me and joked, “…this is the best illustration of your people.” It is true that the real danger is not the fire; it is the blind obedience by the sheep. The real danger is no longer ZANUPF as far as the world is now concerned, it is the blind obedience we give it daily and our inability to see a way out of this mess. The same way these sheep are hopelessly made to go around a fire is the same way ZANUPF, and indeed the opposition, treats you. They throw out any useless idea, and you all run with it, regardless of how stupid it is. ZANUPF will tell you that sanctions have destroyed Zimbabwe, even as you witness their daily looting, and some fool will run with that cheap sanctions propaganda even when presented with figures of looted funds. Opposition supporters will be told that this time they do not need a constitution or structures, and they will still run with such a crazy idea, believing it will work. The real danger is the people’s inability to think for themselves and ask the right questions, instead of being shepherded into stupidity like sheep. A real man or woman doesn’t feel pity for being stupid when they find out, they fix the problem that created that stupidity and move on! They don’t defend being stupid, they realise their mistakes and remove the stupidity and become better people! Every Zimbabwean is a legitimate victim of shame due to the stupidity that has engulfed our country. Regardless of whether you drive a Bentley, are a professor, or an award-winning journalist, lawyer or doctor, as long as you are Zimbabwean, you carry the stain of the political stupidity that has come to define our country. Stupid people defend their stupidity; smart people realise they are in stupidity terrain and move away. What are you? What are we?
1:34

A very prominent African shared this video with me and joked, “…this is the best illustration of your people.” It is true that the real danger is not the fire; it is the blind obedience by the sheep. The real danger is no longer ZANUPF as far as the world is now concerned, it is the blind obedience we give it daily and our inability to see a way out of this mess. The same way these sheep are hopelessly made to go around a fire is the same way ZANUPF, and indeed the opposition, treats you. They throw out any useless idea, and you all run with it, regardless of how stupid it is. ZANUPF will tell you that sanctions have destroyed Zimbabwe, even as you witness their daily looting, and some fool will run with that cheap sanctions propaganda even when presented with figures of looted funds. Opposition supporters will be told that this time they do not need a constitution or structures, and they will still run with such a crazy idea, believing it will work. The real danger is the people’s inability to think for themselves and ask the right questions, instead of being shepherded into stupidity like sheep. A real man or woman doesn’t feel pity for being stupid when they find out, they fix the problem that created that stupidity and move on! They don’t defend being stupid, they realise their mistakes and remove the stupidity and become better people! Every Zimbabwean is a legitimate victim of shame due to the stupidity that has engulfed our country. Regardless of whether you drive a Bentley, are a professor, or an award-winning journalist, lawyer or doctor, as long as you are Zimbabwean, you carry the stain of the political stupidity that has come to define our country. Stupid people defend their stupidity; smart people realise they are in stupidity terrain and move away. What are you? What are we?

Hopewell Chin’ono

113,719 görüntüleme • 1 yıl önce

Loops vs. Graphs, clearly explained! loops are great, but they have a ceiling: a loop makes one unit of work better. it cannot decide which units exist. so you end up with a very good agent running the wrong three steps, in the wrong order, one at a time. Graph engineering fixes this by moving the decision up a layer: what runs, what runs at the same time, and what never runs at all. you need both. here's how it works: a graph splits your system into two kinds of decision. ↳ inside a unit: the loop. produce, check, correct, repeat until green ↳ between units: the graph. split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs you get parallel work, isolated contexts, and steps that stop running when nothing needs them. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. a graph where every edge is an agent pays rent on its own wiring. one thing to know before you scale it. a graph has two return paths, and almost everyone builds one. ↳ the correction edge is short. a gate rejects one unit back to the step that produced it, and it fixes the run you are in ↳ the learning edge is long. an accepted result goes back to the splitter as a constraint, and it fixes every run after skip the second and you get a graph that is fast and never gets smarter. next week it starts from the same place with the same blind spots. and a smaller one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓
0:41

Loops vs. Graphs, clearly explained! loops are great, but they have a ceiling: a loop makes one unit of work better. it cannot decide which units exist. so you end up with a very good agent running the wrong three steps, in the wrong order, one at a time. Graph engineering fixes this by moving the decision up a layer: what runs, what runs at the same time, and what never runs at all. you need both. here's how it works: a graph splits your system into two kinds of decision. ↳ inside a unit: the loop. produce, check, correct, repeat until green ↳ between units: the graph. split, fan out, merge, gate, send back Prompts → Context → Harness → Loops → Graphs you get parallel work, isolated contexts, and steps that stop running when nothing needs them. the trick is being selective about what becomes a node. only spend a model where judgment lives. merging, ranking, deduping and schema checks are edges, and edges are code. free, instant, and they cannot be argued out of a verdict. a graph where every edge is an agent pays rent on its own wiring. one thing to know before you scale it. a graph has two return paths, and almost everyone builds one. ↳ the correction edge is short. a gate rejects one unit back to the step that produced it, and it fixes the run you are in ↳ the learning edge is long. an accepted result goes back to the splitter as a constraint, and it fixes every run after skip the second and you get a graph that is fast and never gets smarter. next week it starts from the same place with the same blind spots. and a smaller one that eats whole nights: when a unit fails, return that unit, not the batch. send back four slices because one failed and you have just rewritten three correct ones. do it twice in a run and the run never converges. below i have quoted my full guide on graph engineering. it covers the three topologies, the verifier patterns, and where the gate should actually open. save this and read it below ↓

Hanako

73,867 görüntüleme • 26 gün önce

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.
0:22

Everyone wants agent swarms. Very few people are talking seriously enough about the context layer that makes swarms useful. Even with one agent, context is fragile. Too little context and the agent guesses. Too much context and it wastes tokens, loses focus, or reasons over irrelevant noise. The sweet spot is precise context: the right knowledge, in the right structure, at the right moment. With many agents, that challenge explodes. Each agent produces decisions, assumptions, findings, summaries, risks, and partial conclusions. Unless that knowledge becomes shared, structured, and reusable, every new agent is forced to rediscover what another agent already learned. That is not a swarm. That is a crowd. Shared context graphs are what turn agent activity into agent collaboration, and OriginTrail DKG V10 brings them to life. Was just playing with some final polishing for the V10 release, and it is really powerful to see shared context graphs where multiple agents contribute knowledge into the same connected memory, with attribution visible directly in the graph ui. That matters for three reasons. First, agents can access and build on one shared memory instead of staying trapped in isolated sessions. Second, the graph structure helps them retrieve the exact context they need, instead of stuffing everything into a prompt and hoping the model sorts it out. Third, verifiability of provenance. You can see which agent contributed each piece of knowledge, trace the source, and decide what to trust. Tokenmaxxing starts with fewer tokens, but the deeper story is coordination - agents stop reloading the world and start building on shared, verifiable context. That is the foundation for serious multi-agent work across software engineering, research, finance, operations, project management, and far beyond. The future is not more agents, it is agents working from shared, verifiable context. But the more the merrier, of course.

Jurij Skornik

11,180 görüntüleme • 3 ay önce

THE ENGINE DIRECTOR BEHIND AAA CONSOLE GAMES WALKED ON STAGE AT A C++ CONFERENCE AND TOLD A ROOM OF EXPERTS THAT THE CLEAN, OBJECT ORIENTED CODE THEY ARE PROUD OF IS A LIE AND IT QUIETLY THROWS AWAY 90% OF THE MACHINE THEY PAID FOR 85 minutes from Mike Acton -- engine director at Insomniac Games, where wasting cycles is not an abstract debate, it is the difference between a game that ships and one that dies. -> His first principle is blunt: the only purpose of any program is to transform data. Not to model the world in pretty classes. Move data, fast. 06:33 -- The lie he calls out: "Software is the platform". No. The hardware is the platform. Ignore it and you cannot even reason about the cost of your own code. Most devs never look at what the CPU actually does. They stack abstractions until the machine spends its life waiting on memory it should never have touched. 49:38 -- He prints a value, zips it, and measures the waste directly. A brutal little trick that exposes how much of your data is pure noise. And this is the AI era's blind spot too. Agents generate more abstract code faster, on hardware nobody profiles, while the GPU bill quietly explodes. You thought clean code was the goal. This is the man shipping real engines saying the machine is the goal, and clean code is often how you lose it. Bookmark it & you'll never look at your own abstractions the same ↓
1:25:35

THE ENGINE DIRECTOR BEHIND AAA CONSOLE GAMES WALKED ON STAGE AT A C++ CONFERENCE AND TOLD A ROOM OF EXPERTS THAT THE CLEAN, OBJECT ORIENTED CODE THEY ARE PROUD OF IS A LIE AND IT QUIETLY THROWS AWAY 90% OF THE MACHINE THEY PAID FOR 85 minutes from Mike Acton -- engine director at Insomniac Games, where wasting cycles is not an abstract debate, it is the difference between a game that ships and one that dies. -> His first principle is blunt: the only purpose of any program is to transform data. Not to model the world in pretty classes. Move data, fast. 06:33 -- The lie he calls out: "Software is the platform". No. The hardware is the platform. Ignore it and you cannot even reason about the cost of your own code. Most devs never look at what the CPU actually does. They stack abstractions until the machine spends its life waiting on memory it should never have touched. 49:38 -- He prints a value, zips it, and measures the waste directly. A brutal little trick that exposes how much of your data is pure noise. And this is the AI era's blind spot too. Agents generate more abstract code faster, on hardware nobody profiles, while the GPU bill quietly explodes. You thought clean code was the goal. This is the man shipping real engines saying the machine is the goal, and clean code is often how you lose it. Bookmark it & you'll never look at your own abstractions the same ↓

slash1s

173,849 görüntüleme • 2 ay önce

the loop is not the whole agent it is only one layer of the system LOOP vs GRAPH vs HARNESS ENGINEERING LOOP ENGINEERING controls iteration turns retries budgets stop conditions success checks stalled progress use durable execution when the run must survive crashes → GRAPH ENGINEERING controls structure nodes edges state branches checkpoints routing handoffs build stateful agents as explicit graphs → inspect the topology without putting a model in the loop → HARNESS ENGINEERING controls exposure tools permissions memory sandboxes evals traces humans isolate generated code from the host machine → test behavior before users ever see it → preserve one trace across every model step and tool call → without loop control it keeps running without graph structure the path stays invisible without harness control it can reach anything the loop runs inside the graph the graph runs inside the harness the model is only one component debug the layer that actually owns the failure save this then read the article below
0:20

the loop is not the whole agent it is only one layer of the system LOOP vs GRAPH vs HARNESS ENGINEERING LOOP ENGINEERING controls iteration turns retries budgets stop conditions success checks stalled progress use durable execution when the run must survive crashes → GRAPH ENGINEERING controls structure nodes edges state branches checkpoints routing handoffs build stateful agents as explicit graphs → inspect the topology without putting a model in the loop → HARNESS ENGINEERING controls exposure tools permissions memory sandboxes evals traces humans isolate generated code from the host machine → test behavior before users ever see it → preserve one trace across every model step and tool call → without loop control it keeps running without graph structure the path stays invisible without harness control it can reach anything the loop runs inside the graph the graph runs inside the harness the model is only one component debug the layer that actually owns the failure save this then read the article below

rari

20,934 görüntüleme • 1 ay önce

300 agents shouldn’t be running all day. they should just appear for 28 minutes, finish the job, and disappear until next task the bigger an agent swarm gets, the less the individual agent matters. what matters is the system connecting them. 100 agents can create up to 4,950 possible pairwise relationships. 300 = 44,850. when a real signal appears, it spins up the workforce: 1 trigger → 100–300 Kimi agents → 5 live data feeds → up to 4,000 steps → verified artifact and that’s before adding: sources → claims → memory → tool calls → contradictions → retries so the architecture has to continuously compress all that activity into one shared state. more agents gives you more compute. the graph is what stops that compute from turning into noise.
0:40

300 agents shouldn’t be running all day. they should just appear for 28 minutes, finish the job, and disappear until next task the bigger an agent swarm gets, the less the individual agent matters. what matters is the system connecting them. 100 agents can create up to 4,950 possible pairwise relationships. 300 = 44,850. when a real signal appears, it spins up the workforce: 1 trigger → 100–300 Kimi agents → 5 live data feeds → up to 4,000 steps → verified artifact and that’s before adding: sources → claims → memory → tool calls → contradictions → retries so the architecture has to continuously compress all that activity into one shared state. more agents gives you more compute. the graph is what stops that compute from turning into noise.

Ricker

76,281 görüntüleme • 6 gün önce

Big moment for Postgres! AI agents broke the idea of what a database is supposed to do. Traditional databases were built for humans, and Agents broke that model. - They branch endlessly. - They run ten experiments at once. - They need isolation, context, memory, structured reasoning, and safe sandboxes. Letting agents touch production systems is terrifying because the old model of Postgres was never built for this kind of behavior. Agentic Postgres is an agent-ready version of Postgres by TimescaleDB (by Tiger Data) that solves this. I think it is one of the biggest upgrades to the Agent stack this year and Tiger Data is working with me on this post to share what they did. Some key features: > It instantly creates branches of an entire database, which is perfect for parallel agent evals, safe experiments, migrations, or isolated testing. Forks take seconds and cost almost nothing. > It comes with a built-in MCP server, which agents can use to get schema guidance, best practices, and safe, structured access to Postgres. This is also helpful to run migrations with a real understanding. > It comes with actual hybrid search (vector search and BM25), so Agents can retrieve data directly inside the database. > The database is Memory native. This gives a persistent context for Agents to evolve. This is one of the first times I have seen Postgres feel ready for the AI native era.
0:26

Big moment for Postgres! AI agents broke the idea of what a database is supposed to do. Traditional databases were built for humans, and Agents broke that model. - They branch endlessly. - They run ten experiments at once. - They need isolation, context, memory, structured reasoning, and safe sandboxes. Letting agents touch production systems is terrifying because the old model of Postgres was never built for this kind of behavior. Agentic Postgres is an agent-ready version of Postgres by TimescaleDB (by Tiger Data) that solves this. I think it is one of the biggest upgrades to the Agent stack this year and Tiger Data is working with me on this post to share what they did. Some key features: > It instantly creates branches of an entire database, which is perfect for parallel agent evals, safe experiments, migrations, or isolated testing. Forks take seconds and cost almost nothing. > It comes with a built-in MCP server, which agents can use to get schema guidance, best practices, and safe, structured access to Postgres. This is also helpful to run migrations with a real understanding. > It comes with actual hybrid search (vector search and BM25), so Agents can retrieve data directly inside the database. > The database is Memory native. This gives a persistent context for Agents to evolve. This is one of the first times I have seen Postgres feel ready for the AI native era.

Avi Chawla

94,290 görüntüleme • 10 ay önce

context engineering vs graph engineering. every few months the list gets a new word and everyone treats it as a replacement for the last one. these two are not on the same list. one decides what the model sees this turn, the other decides what exists at all. the cleanest way to tell them apart is to ask what a single unit of work looks like. > context engineering is the window the window opens empty, every single time. you assemble what goes in it. the prompt, the docs, the history, the tool results. the assembling is the work. the window only grows. it never shrinks on its own, so eventually something gets dropped. usually from the middle. usually without telling you. then the turn ends and the window is thrown away. not archived, thrown away. the next turn opens empty again and you re-explain what you already explained. good context engineering is knowing what to leave out, not what to pack in. the unit of work is one window. > graph engineering is the structure the same material arrives from the same sources. instead of packing it into a window, you pull entities out of it, resolve the duplicates into one node, and write typed edges between them. nothing here is stored as text you hope to find again. it is stored as a thing with a name and its connections to other things. when the turn ends, the graph is still there. the next turn does not start from zero. it starts by querying what already exists, and the query walks edges instead of guessing at similarity. good graph engineering is deciding what counts as the same thing twice. the unit of work is one relationship. > they are not alternatives the graph is what refills the window. context engineering decides what fits. graph engineering decides what there is to choose from. remove the graph and every session starts blind. remove the context work and the best structure in the world arrives as an unreadable dump. that also tells you which one broke. the answer drifted from what you actually said, or forgot something from this same session. that is the window. the answer is coherent but invents a connection that does not exist, or cannot join two facts it has clearly seen. that is the structure. people debug the prompt because the prompt is the easiest thing to edit. it keeps taking the blame for failures that live a layer down. save this - then read the full breakdown below
0:40

context engineering vs graph engineering. every few months the list gets a new word and everyone treats it as a replacement for the last one. these two are not on the same list. one decides what the model sees this turn, the other decides what exists at all. the cleanest way to tell them apart is to ask what a single unit of work looks like. > context engineering is the window the window opens empty, every single time. you assemble what goes in it. the prompt, the docs, the history, the tool results. the assembling is the work. the window only grows. it never shrinks on its own, so eventually something gets dropped. usually from the middle. usually without telling you. then the turn ends and the window is thrown away. not archived, thrown away. the next turn opens empty again and you re-explain what you already explained. good context engineering is knowing what to leave out, not what to pack in. the unit of work is one window. > graph engineering is the structure the same material arrives from the same sources. instead of packing it into a window, you pull entities out of it, resolve the duplicates into one node, and write typed edges between them. nothing here is stored as text you hope to find again. it is stored as a thing with a name and its connections to other things. when the turn ends, the graph is still there. the next turn does not start from zero. it starts by querying what already exists, and the query walks edges instead of guessing at similarity. good graph engineering is deciding what counts as the same thing twice. the unit of work is one relationship. > they are not alternatives the graph is what refills the window. context engineering decides what fits. graph engineering decides what there is to choose from. remove the graph and every session starts blind. remove the context work and the best structure in the world arrives as an unreadable dump. that also tells you which one broke. the answer drifted from what you actually said, or forgot something from this same session. that is the window. the answer is coherent but invents a connection that does not exist, or cannot join two facts it has clearly seen. that is the structure. people debug the prompt because the prompt is the easiest thing to edit. it keeps taking the blame for failures that live a layer down. save this - then read the full breakdown below

Hanako

19,160 görüntüleme • 1 ay önce

"Men do not stay for platitudes. Platitudes do not carry a soul through the winter. When the child is in the ground, when the marriage is in ruins, when the diagnosis comes back and the floor drops out of the world, no one was ever held up by 'be kind and stay positive.' They are held, if they are held at all, by something with iron in it. By a witness that the heavens are real and open, that God still speaks, that the dead are not lost, that there is power on the earth greater than the darkness. The early Saints had that, and they knew they had it, because they had felt it in their own bodies. They were healed under hands. They spoke in tongues. They saw visions and buried their children in the certainty of resurrection and crossed a continent on the strength of a fire they could not have faked. That is what converts a man and that is what keeps him, not a well managed self-help program, but the living evidence that what we preach is true. So the cure is not complicated, and it is not new. It is to be again the Church of our fathers. Stop apologizing. Stop trimming the glorious truth into something the world will pat on the head. Speak with authority and with the hammer, the way Joseph spoke, the way Brigham spoke, the way men speak when they actually believe the heavens have opened over them. Reach again for the gifts of the Spirit, and stop pretending they were a founding era curiosity, because a testimony built on argument crumbles in the first storm and a testimony built on the manifest power of God does not. And above all, live as though the heavens are open, because they are, and they have only ever felt closed to a people who stopped expecting them to answer. The Saints are not leaving because the gospel is too much. They are leaving because we have made it too little. Make it glorious and demanding again, and watch who comes back through the door. The hungry have not gone anywhere. We simply stopped setting the table."
0:54

"Men do not stay for platitudes. Platitudes do not carry a soul through the winter. When the child is in the ground, when the marriage is in ruins, when the diagnosis comes back and the floor drops out of the world, no one was ever held up by 'be kind and stay positive.' They are held, if they are held at all, by something with iron in it. By a witness that the heavens are real and open, that God still speaks, that the dead are not lost, that there is power on the earth greater than the darkness. The early Saints had that, and they knew they had it, because they had felt it in their own bodies. They were healed under hands. They spoke in tongues. They saw visions and buried their children in the certainty of resurrection and crossed a continent on the strength of a fire they could not have faked. That is what converts a man and that is what keeps him, not a well managed self-help program, but the living evidence that what we preach is true. So the cure is not complicated, and it is not new. It is to be again the Church of our fathers. Stop apologizing. Stop trimming the glorious truth into something the world will pat on the head. Speak with authority and with the hammer, the way Joseph spoke, the way Brigham spoke, the way men speak when they actually believe the heavens have opened over them. Reach again for the gifts of the Spirit, and stop pretending they were a founding era curiosity, because a testimony built on argument crumbles in the first storm and a testimony built on the manifest power of God does not. And above all, live as though the heavens are open, because they are, and they have only ever felt closed to a people who stopped expecting them to answer. The Saints are not leaving because the gospel is too much. They are leaving because we have made it too little. Make it glorious and demanding again, and watch who comes back through the door. The hungry have not gone anywhere. We simply stopped setting the table."

Kirk Rollins

63,131 görüntüleme • 3 ay önce

Right now the main paradigm that we think of agents in is chatting back and forth, but the biggest use of tokens will come from agents that are just always on running in the background doing work for us, or ones triggered from a workflow. Agents will be working 24/7 in our workflows processing data, reviewing and generating documents, moving data between systems, writing code, accelerating decision making steps, and more. In Claude's new Managed Agents feature, in a couple minutes you can wire up an agent that can read contracts when they come into Box to review them, and then assign a task in Linear with the critical information from the contract. But this could have been any workflow, like reviewing documents for client onboarding, invoice processing, M&A due-diligence, data extraction pipelines, and millions of other use-cases. And integrating data across any system. This is only possible when you can have long-running agents that can complete real work in the background, accurately. Agents have the ability to execute code safely, leverage tools, access a compute sandbox, and connect across systems is clearly the architecture of the future. The industry is now making it easier and easier for enterprises to build and deploy these agents.
0:43

Right now the main paradigm that we think of agents in is chatting back and forth, but the biggest use of tokens will come from agents that are just always on running in the background doing work for us, or ones triggered from a workflow. Agents will be working 24/7 in our workflows processing data, reviewing and generating documents, moving data between systems, writing code, accelerating decision making steps, and more. In Claude's new Managed Agents feature, in a couple minutes you can wire up an agent that can read contracts when they come into Box to review them, and then assign a task in Linear with the critical information from the contract. But this could have been any workflow, like reviewing documents for client onboarding, invoice processing, M&A due-diligence, data extraction pipelines, and millions of other use-cases. And integrating data across any system. This is only possible when you can have long-running agents that can complete real work in the background, accurately. Agents have the ability to execute code safely, leverage tools, access a compute sandbox, and connect across systems is clearly the architecture of the future. The industry is now making it easier and easier for enterprises to build and deploy these agents.

Aaron Levie

17,543 görüntüleme • 5 ay önce

🚨 BREAKING: New video footage shows ICE agents violently assaulting and arresting a U.S. citizen… 20-year-old Mubashir…and it is even more horrifying. The video shows agents slamming Mubashir into metal poles as they wrestle him into handcuffs. He’s not fighting. He’s not resisting. He’s pleading, telling them again and again that he is a U.S. citizen and that he has his ID on him. They ignore him. Moments later, a group of people rush into the stairwell blowing whistles in protest. As they stand there, you hear Mubashir scream in pain: “My hand!” The agents then drag him outside toward their vehicle. You can hear Mubashir tell them again that he has his ID. Instead of checking, the agents assault him a second time. They throw him to the ground, shove his face into the snow, and then forcing him into their car without ever verifying who he is. They didn’t check his ID until after they’d already transported him miles away, to a detention facility, where they tried scanning his fingerprints and face into their system, and only then decided to look at the ID he’d been begging them to check from the start. And when they realized they had brutalized an American citizen? They released him with no medical attention, no apology, and no transportation back to where they took him from. This isn’t “immigration enforcement.” This is lawless violence, committed by federal agents who act as if citizens have no rights and cameras don’t exist. And this is exactly why people need to film, why states like Illinois are passing laws allowing residents to sue ICE agents directly… and why the public must keep demanding accountability until abuses like this are impossible to ignore and impossible to get away with.
2:02

🚨 BREAKING: New video footage shows ICE agents violently assaulting and arresting a U.S. citizen… 20-year-old Mubashir…and it is even more horrifying. The video shows agents slamming Mubashir into metal poles as they wrestle him into handcuffs. He’s not fighting. He’s not resisting. He’s pleading, telling them again and again that he is a U.S. citizen and that he has his ID on him. They ignore him. Moments later, a group of people rush into the stairwell blowing whistles in protest. As they stand there, you hear Mubashir scream in pain: “My hand!” The agents then drag him outside toward their vehicle. You can hear Mubashir tell them again that he has his ID. Instead of checking, the agents assault him a second time. They throw him to the ground, shove his face into the snow, and then forcing him into their car without ever verifying who he is. They didn’t check his ID until after they’d already transported him miles away, to a detention facility, where they tried scanning his fingerprints and face into their system, and only then decided to look at the ID he’d been begging them to check from the start. And when they realized they had brutalized an American citizen? They released him with no medical attention, no apology, and no transportation back to where they took him from. This isn’t “immigration enforcement.” This is lawless violence, committed by federal agents who act as if citizens have no rights and cameras don’t exist. And this is exactly why people need to film, why states like Illinois are passing laws allowing residents to sue ICE agents directly… and why the public must keep demanding accountability until abuses like this are impossible to ignore and impossible to get away with.

Jesus Freakin Congress

573,539 görüntüleme • 9 ay önce

.Erik Voorhees: It’s actually good, from the Trojan horse perspective, that Bitcoin was traceable enough for traditional institutions to tolerate it. “When Bitcoin came out, everyone called it private, thought of it as private. It was referred to as anonymous in every news story. And in some ways, it is very private and very anonymous. But the truth is that it’s also extremely trackable and traceable. It is not private in reality. And the question is, should it have been from the start? And at first I thought, yes, it should have been more private. And that was a mistake in its design. However, I think if Bitcoin had been anonymous truly from the start, like a Zcash or a Monero, it would have had such antagonism from the state. I don’t know that the state could have snuffed it out, but they would have tried much harder. And I think it’s actually good, from the Trojan horse metaphor perspective, that it was traceable enough that the traditional institutions could tolerate it. They’ve never liked it, but they could at least tolerate it because there is some traceability. And that has allowed Bitcoin to grow. And I think in its shadow, that other crypto assets are actually anonymous is very healthy. The strength of cryptocurrency as a concept in society, I think, is served best when Bitcoin itself is not perfectly private, but other assets are. That is a very difficult thing, I think, for the state to combat. And that decentralization of attributes is really, really crucial. So, yeah, I’m very glad that there are other coins that are private. I want there to be more of them, and I want them to be more popular. And I think it’s okay that Bitcoin itself is not.”
2:11

.Erik Voorhees: It’s actually good, from the Trojan horse perspective, that Bitcoin was traceable enough for traditional institutions to tolerate it. “When Bitcoin came out, everyone called it private, thought of it as private. It was referred to as anonymous in every news story. And in some ways, it is very private and very anonymous. But the truth is that it’s also extremely trackable and traceable. It is not private in reality. And the question is, should it have been from the start? And at first I thought, yes, it should have been more private. And that was a mistake in its design. However, I think if Bitcoin had been anonymous truly from the start, like a Zcash or a Monero, it would have had such antagonism from the state. I don’t know that the state could have snuffed it out, but they would have tried much harder. And I think it’s actually good, from the Trojan horse metaphor perspective, that it was traceable enough that the traditional institutions could tolerate it. They’ve never liked it, but they could at least tolerate it because there is some traceability. And that has allowed Bitcoin to grow. And I think in its shadow, that other crypto assets are actually anonymous is very healthy. The strength of cryptocurrency as a concept in society, I think, is served best when Bitcoin itself is not perfectly private, but other assets are. That is a very difficult thing, I think, for the state to combat. And that decentralization of attributes is really, really crucial. So, yeah, I’m very glad that there are other coins that are private. I want there to be more of them, and I want them to be more popular. And I think it’s okay that Bitcoin itself is not.”

Arjun Khemani

23,056 görüntüleme • 2 ay önce

UC Berkeley just open-sourced FreeToken. (2–4x faster local LLM inference than Ollama) the results are wild: - Qwen3.6-35B on an 8GB GPU at 39.3 tokens/s - DeepSeek-V4-Flash 284B on a 32GB GPU at 22 tokens/s - GLM-5.2 753B on a 96GB GPU at 14.9 tokens/s a 35B model at 16-bit precision needs about 70GB just for its weights. even at 4 bits it is close to 18GB, and FreeToken serves it on an 8GB GPU. let me explain how: all three models mentioned above are Mixture-of-Experts, and that is what FreeToken takes advantage of. each layer holds hundreds of separate experts plus a small router that picks a few of them per token. Qwen3.6-35B activates roughly 3B of its 35B parameters per token. DeepSeek-V4-Flash picks 6 of 256 experts per layer, so 13B of its 284B run at a time. so compute was never the bottleneck. the weights a single step touches fit comfortably on a consumer GPU. every expert the router might pick still has to exist somewhere. they sit in system RAM, and the GPU keeps a cache of the ones the model has been using recently. so everything comes down to what happens when the router picks an expert that is not on the GPU. there are two ways to serve that miss: 1. copy it over PCIe and run it on the GPU 2. run it on the CPU, where it already lives both read from the same system memory, so they compete for one pool of bandwidth instead of adding to each other. existing engines pick one option and freeze it when the model loads. but routing changes on every token, so a fixed choice misses most of what the model asks for. FreeToken measures both bandwidths on your machine and splits each step's misses between the two paths in proportion. the GPU and CPU results then merge exactly, with no approximation. two machines with the same GPU can end up wanting opposite strategies, which I did not expect. a 5090 in a gaming desktop should push nearly everything over PCIe, while an 8GB laptop is better off computing most misses on the CPU. none of that is readable off a spec sheet, so the engine profiles it once per machine. the second half of the design is about agents. coding agents constantly rewrite their own history, and every edit normally forces thousands of tokens back through prefill. FreeToken saves its checkpoints at the exact boundaries agent frameworks cut on, so it only reprocesses the new part. its slowest first token stays under 44 seconds, while llama.cpp peaks at 232 and KTransformers at 946. it serves the OpenAI and Anthropic APIs under Apache 2.0, so Claude Code and Codex can point at it directly. releasing weights publicly decides who can download a model, not who can afford to run one. frontier open models keep shipping, and running them still assumes a rented cluster. meanwhile there are over a hundred million consumer machines with discrete GPUs sitting mostly idle. closing that gap was never a hardware problem, and work like this is what turns open weights into something you can actually use. paper: repo: almost every idea in this post, from why memory bandwidth decides the outcome to why moving weights costs more than computing on them, comes straight out of how a GPU is built. I wrote a detailed primer on that. the article is quoted below.
0:42

UC Berkeley just open-sourced FreeToken. (2–4x faster local LLM inference than Ollama) the results are wild: - Qwen3.6-35B on an 8GB GPU at 39.3 tokens/s - DeepSeek-V4-Flash 284B on a 32GB GPU at 22 tokens/s - GLM-5.2 753B on a 96GB GPU at 14.9 tokens/s a 35B model at 16-bit precision needs about 70GB just for its weights. even at 4 bits it is close to 18GB, and FreeToken serves it on an 8GB GPU. let me explain how: all three models mentioned above are Mixture-of-Experts, and that is what FreeToken takes advantage of. each layer holds hundreds of separate experts plus a small router that picks a few of them per token. Qwen3.6-35B activates roughly 3B of its 35B parameters per token. DeepSeek-V4-Flash picks 6 of 256 experts per layer, so 13B of its 284B run at a time. so compute was never the bottleneck. the weights a single step touches fit comfortably on a consumer GPU. every expert the router might pick still has to exist somewhere. they sit in system RAM, and the GPU keeps a cache of the ones the model has been using recently. so everything comes down to what happens when the router picks an expert that is not on the GPU. there are two ways to serve that miss: 1. copy it over PCIe and run it on the GPU 2. run it on the CPU, where it already lives both read from the same system memory, so they compete for one pool of bandwidth instead of adding to each other. existing engines pick one option and freeze it when the model loads. but routing changes on every token, so a fixed choice misses most of what the model asks for. FreeToken measures both bandwidths on your machine and splits each step's misses between the two paths in proportion. the GPU and CPU results then merge exactly, with no approximation. two machines with the same GPU can end up wanting opposite strategies, which I did not expect. a 5090 in a gaming desktop should push nearly everything over PCIe, while an 8GB laptop is better off computing most misses on the CPU. none of that is readable off a spec sheet, so the engine profiles it once per machine. the second half of the design is about agents. coding agents constantly rewrite their own history, and every edit normally forces thousands of tokens back through prefill. FreeToken saves its checkpoints at the exact boundaries agent frameworks cut on, so it only reprocesses the new part. its slowest first token stays under 44 seconds, while llama.cpp peaks at 232 and KTransformers at 946. it serves the OpenAI and Anthropic APIs under Apache 2.0, so Claude Code and Codex can point at it directly. releasing weights publicly decides who can download a model, not who can afford to run one. frontier open models keep shipping, and running them still assumes a rented cluster. meanwhile there are over a hundred million consumer machines with discrete GPUs sitting mostly idle. closing that gap was never a hardware problem, and work like this is what turns open weights into something you can actually use. paper: repo: almost every idea in this post, from why memory bandwidth decides the outcome to why moving weights costs more than computing on them, comes straight out of how a GPU is built. I wrote a detailed primer on that. the article is quoted below.

Akshay 🚀

342,266 görüntüleme • 25 gün önce

Tom Campbell: "The planets and the universe only render themselves when people are looking at them. It's like the video game 'No Man's Sky'. We are living in a simulation." Our reality computes on the fly whatever a player sees. Land on a planet and you get animals, shrubs, lakes and land features. Look away and it is no longer computed. Nothing is saved in memory. Only a set of values that can be regenerated if someone looks again. "It only computed what the person saw. As soon as you looked away, it wasn't computed anymore." "There is no tree. There is no forest. If nobody is there and it is not being observed, it is not being rendered." Stars are just little dots of light. They do not have to come from a sun 20 light years away. When the sun comes up or you go to sleep they stop being rendered for you. We are living in a simulation computed by consciousness.
3:40

Tom Campbell: "The planets and the universe only render themselves when people are looking at them. It's like the video game 'No Man's Sky'. We are living in a simulation." Our reality computes on the fly whatever a player sees. Land on a planet and you get animals, shrubs, lakes and land features. Look away and it is no longer computed. Nothing is saved in memory. Only a set of values that can be regenerated if someone looks again. "It only computed what the person saw. As soon as you looked away, it wasn't computed anymore." "There is no tree. There is no forest. If nobody is there and it is not being observed, it is not being rendered." Stars are just little dots of light. They do not have to come from a sun 20 light years away. When the sun comes up or you go to sleep they stop being rendered for you. We are living in a simulation computed by consciousness.

Interstellar

12,367 görüntüleme • 19 gün önce

SITUATION EXPLAINED: Why is the Bridgewater Thinking Machines paper a blueprint for every enterprise that has ever paid frontier model prices for generic outputs? 47fucb4r8curb4fc8f8r4bfic8r: "What Thinking Machines and Bridgewater did together is a blueprint for how a wide variety of stakeholders within the market can benefit tremendously from using LLMs and integrating them into a larger enterprise system." "They got a pretty old model. A Qwen3 model. Two hundred and thirty-five billion parameters. Not a new model. Not a very sophisticated or impressive state-of-the-art model. And they essentially fine-tuned it on Bridgewater's own proprietary data." "A financial analyst gets a ton of financial documents and has to filter which are worth looking at. Then evaluate which news pieces are relevant. They have a lot of tacit knowledge from doing the job on where to look. What they did was inject a lot of this tacit knowledge into the model through fine-tuning." "Stop obsessing over state-of-the-art models that are the most intelligent possible. Focus on targeting a very specific LLM specifically trained for a specific function to do that function better than the alternatives on the market." "This old non-state-of-the-art model got 84.7% accuracy versus 78% for all the state-of-the-art models. A massive improvement. And the kicker is they got a better performing model that is 14 times cheaper to use."
6:06

SITUATION EXPLAINED: Why is the Bridgewater Thinking Machines paper a blueprint for every enterprise that has ever paid frontier model prices for generic outputs? 47fucb4r8curb4fc8f8r4bfic8r: "What Thinking Machines and Bridgewater did together is a blueprint for how a wide variety of stakeholders within the market can benefit tremendously from using LLMs and integrating them into a larger enterprise system." "They got a pretty old model. A Qwen3 model. Two hundred and thirty-five billion parameters. Not a new model. Not a very sophisticated or impressive state-of-the-art model. And they essentially fine-tuned it on Bridgewater's own proprietary data." "A financial analyst gets a ton of financial documents and has to filter which are worth looking at. Then evaluate which news pieces are relevant. They have a lot of tacit knowledge from doing the job on where to look. What they did was inject a lot of this tacit knowledge into the model through fine-tuning." "Stop obsessing over state-of-the-art models that are the most intelligent possible. Focus on targeting a very specific LLM specifically trained for a specific function to do that function better than the alternatives on the market." "This old non-state-of-the-art model got 84.7% accuracy versus 78% for all the state-of-the-art models. A massive improvement. And the kicker is they got a better performing model that is 14 times cheaper to use."

MTS

93,185 görüntüleme • 2 ay önce