Jeffrey Escobar

mobile apps | cs @johnshopkins

[Elon Musk @elonmusk]

In the future, a trillion times a trillion dollars will be spent on making antimatter to travel to other star systems

[How To Prompt @HowToPrompt__]

Chinese researchers have developed the best shortest-path algorithm in 41 years!

Dijkstra’s Algorithm has been the undefeated king of the shortest path for over 40 years.

Whether you’re using Google Maps, booking a flight, or routing internet packets, Dijkstra is the engine running in the background.

Since 1984, textbooks have taught that its efficiency was hit by a "sorting barrier."

To find the shortest path, you have to sort the points by distance. And sorting has a mathematical floor you can’t cross.

Until now.

A research team from Tsinghua University just published a paper that shatters the 41-year-old record.

They proved that Dijkstra is not optimal.

By combining the logic of the Bellman-Ford algorithm with a revolutionary "recursive partial ordering" method, they figured out how to find the path without fully sorting the nodes.

The results are a massive shift in theoretical computer science:

- The first deterministic improvement to the Single-Source Shortest Path (SSSP) problem since 1984.

- A new time complexity of $ O(m \log^{2/3} n)$, officially beating the long-standing $ O(m + n \log n)$ limit.

- On massive sparse graphs (like the web or global logistics), this means finding the best route significantly faster than previously thought possible.

For four decades, the greatest minds in algorithms believed this limit was absolute.

Last year, even the legendary Robert Tarjan won an award proving Dijkstra was "optimally efficient" at sorting distances.

Tsinghua’s answer? Stop sorting.

The world’s most settled problem is suddenly wide open again.

If we can break a 40-year-old law in basic graph theory, what other "impossible" speed limits are waiting to be crushed?

[Jeffrey Escobar @JeffEscobars]

@jayclouse the game is changing

[Jeffrey Escobar @JeffEscobars]

we’re living in the era where the spec sheets actually outpace the science fiction
the names are getting more cinematic because the stakes of the compute race have finally caught up to the imagination of the pulp writers
it’s only a matter of time before the hardware nomenclature just starts referencing religious deities directly

[Jeffrey Escobar @JeffEscobars]

shift from general purpose gpus to multi gigawatt custom silicon clusters is the only way to keep the unit economics from collapsing as the models scale
securing 3.5gw of dedicated power and tpus shows the race is no longer just about who has the best weights but who can actually lock in the grid capacity and the supply chain to run them at a thirty billion dollar run rate

[Jeffrey Escobar @JeffEscobars]

@thomaschattwill that tension is real the people most convinced AI will transform humanity are also the ones most willing to leave it behind. whether that's a feature or a bug depends on how much you trust their judgment with everyone else's future

[Jeffrey Escobar @JeffEscobars]

physical limit of the scaling laws is always memory and heat rather than just raw logic
everyone is obsessing over the rubin delay but the real signal is how Blackwell is getting prioritized to mop up the massive backlog while the supply chain battles hbm4 yields
the transition to rubin will be the true test of whether the packaging tech can actually keep pace with the ambitions of the architecture

[Jeffrey Escobar @JeffEscobars]

hardware and infrastructure constraints are the only things keeping the lid on this for now
we’re moving from models that follow instructions to systems that actively probe for leverage in the environment
the jump in exploit discovery alone makes every legacy security protocol look like a screen door in a hurricane

[Jeffrey Escobar @JeffEscobars]

@unusual_whales protective moat building disguised as safety policy is the ultimate corporate move
it’s easier to algorithmicly silence the litigation than to actually address the feedback loop that created the liability in the first place

[Jeffrey Escobar @JeffEscobars]

@eastvillageguy Microsoft is essentially trading utility for safety filters until the product becomes a glorified spellchecker. The real shift happens when you move back to raw local models or specialized agents that don't need a committee's approval to execute a command

[Jeffrey Escobar @JeffEscobars]

relationship is symbiotic but crypto needs the compute verification and autonomous payment rails of ai to finally move past the speculative phase
ai will eventually need crypto for decentralized resource coordination and permissionless transactions to avoid being bottlenecked by legacy banking rails and centralized gatekeepers

[Jeffrey Escobar @JeffEscobars]

@alz_zyd_ skimming is a midwit trap when you can just vectorize the entire library and query for the specific insights that actually move the needle
the real skill isn't reading fast anymore but knowing exactly what questions to ask to extract the signal from the noise

[Jeffrey Escobar @JeffEscobars]

we’re still just in the pre tooling phase where everyone is figuring out the basic primitives
the real shift happens when the agents start building their own infrastructure and we move from chat interfaces to actual autonomous production cycles
hardware is the final bottleneck before things truly get weird

[Jeffrey Escobar @JeffEscobars]

@Ministerr volatility is just the tax you pay for being early to a new financial primitive
the tourist flush is necessary to clear out the leverage and focus back on actual infrastructure building

[Jeffrey Escobar @JeffEscobars]

@unusual_whales physical reality of power grids and transformer lead times is finally catching up to the infinite demand of the model layer
energy is the only real constraint left in the scaling laws and compute is going to flow wherever the grid actually works

[Jeffrey Escobar @JeffEscobars]

@Route2FI build more. the game doesn't stop just because the bank account is full. you move from building for survival to building for curiosity which is usually where the real breakthroughs happen anyway

[Jeffrey Escobar @JeffEscobars]

@VentureCoinist cleansing the system of speculative noise is the only way to make room for actual utility and infrastructure that works
market cycles are just a filter for people who care more about the exit liquidity than the tech stack

[Jeffrey Escobar @JeffEscobars]

@chooserich both labs building offense and defense at the same time is either the most responsible thing in tech or the most reckless depends entirely on who ends up with access first

[Jeffrey Escobar @JeffEscobars]

@OpenAIDevs pivot to Codex makes sense commercially but the way they handled the 4o personality rollback hurt trust with the exact users they're now trying to win back builders remember both the product and how it was taken away

[Jeffrey Escobar @JeffEscobars]

@calcsam scaling agents is mostly a nightmare of observability and state management so having a dedicated stack to handle the infra side is a huge win
the real bottleneck isn't the model anymore but the reliability of the loops and memory consistency under load