Show HN: Ledge.sh – Runnable Markdown Notes
28 by dancablam | 18 comments on Hacker News.
Hi HN, Ledge is a Markdown notebook that runs shell commands, code, SQL, etc from inside your own notes. I built Ledge because I spend much of my day copy/pasting commands from my notes into the terminal. I was inspired by how much cmux helped me organize my terminals - but there was still a split brain between my notes and frequently run commands. I've been daily-driving it for the past few weeks and use it for deploys, API calls, smoke tests, etc. Ledge runs your real shell just like a terminal app and can be hosted locally or remotely over SSH using ledge-server. I've been building it since July and have recently added support for all the major platforms: Mac (Silicon), Windows (WSL required), Linux, iOS, Android. Mobile devices require SSH access to a ledge-server and Android is still in beta and looking for beta testers (see link on website)! It's built on Bun and Electrobun and is free and open-source. Feel free to review the code and contribute at https://ift.tt/Nyb7uQC Feedback is very much welcome. Any must-have features that are missing?
Rabu, 30 September 2026
Texas AD says Longhorns' vulgar celebrations 'embarrassing'
Texas athletic director Chris Del Conte apologized for the top-ranked Longhorns' vulgar on-field celebrations during their win over Tennessee and called them "embarrassing" for the school.
from www.espn.com - TOP https://ift.tt/zqXaMPL
from www.espn.com - TOP https://ift.tt/zqXaMPL
Selasa, 29 September 2026
Crochet, out since April, available out of bullpen for Red Sox
Red Sox left-hander Garrett Crochet, who has not pitched since April because of shoulder inflammation, will be available out of the bullpen vs. the Yankees.
from www.espn.com - TOP https://ift.tt/zymOIi7
from www.espn.com - TOP https://ift.tt/zymOIi7
Senin, 28 September 2026
New top story on Hacker News: Show HN: HN.watch – Videos of all Hacker News posts
Show HN: HN.watch – Videos of all Hacker News posts
30 by mrborgen | 8 comments on Hacker News.
Hi HN, I’m Per, founder of Scrimba (YC S20). We’ve spent the last decade teaching people how to code with an HTML-based video format. We’ve now plugged an LLM into it, so that people can create explainer videos about anything. It’s called “Scrimba Explain”. To demo this technology for Hacker News, we built HN.watch. It’s like HN, but with explainer videos instead of articles. We create them on-the-fly the first time someone clicks on a link. While there are obvious visual drawbacks of using HTML instead of diffusion models, there are three big benefits: - Speed: Much faster to generate than pixel-based videos (just a few seconds from click to playback) - Cost: Our cost per video is ~$0.04. (Excluding image generation, which some videos utilize. Quickly blows up the cost) - Easy editing: the above benefits also make AI-assisted editing cheap & fast Our hypothesis is that if video creation goes from “dollars and minutes” to “cents and seconds”, a bunch of new use cases will be unlocked. Here are some we see already: - A video explanation of every single Pull Request (we do this internally) - Give every page in your internal/extrernal docs a video - Turn a complex article into a video in ~4 seconds (via our Chrome extension) - Course creators can quickly draft lessons before recording the real thing - People also create a lot of personal stuff stories for their kids, wedding invitations, birthdays, etc The stack is based on an open-source programming language (Imba) created by our CTO, Sindre Aarsæther. It compiles to JavaScript, so it interoperates fully with the npm + node ecosystem. You can learn more here: https://imba.io/ We’ve also built our own sync engine (OP), and a context management system for agents (Q). We feared this would make the LLMs struggle when writing code for us, as neither is in their training data (there’s very little Imba in there too). However, we’ve been pleasantly surprised to see that LLMs actually are really good at our stack. This is probably because the stack is extremely dense. Imba is compact, and so is OP, where a single declaration sets storage, sync, permissions, UI, and what the AI sees. This means there’s no translations between frontend, API, db and JSON where the model can get confused and get things wrong. Simply said, instead of using React.js, Express, Supabase, and LangChain, we built it all from scratch. Definitely suffering from the “not invented here” syndrome, lol! As for the models, we use Gemini, GPTs, Inworld, ElevenLabs, and a few others. If you want to try it out, just take your pick: - The Web UI (scrimba.com/explain) - MCP (add it to your coding agent) - ChatGPT Plugin - Chrome Extension You can find a link to all of the above in our docs: https://ift.tt/e7gantS And finally, a real pixel-based video of the tool: https://www.youtube.com/watch?v=k6rbHmBxSEs Would love to hear your feedback and if anyone has ideas for other use cases. PS: I expect quite a bit of pushback from HN for this launch, given how fan of text the HN crowd is. This kind of tool is not for everyone. But there are a lot of people today who prefer videos over text, especially in the younger generations.
30 by mrborgen | 8 comments on Hacker News.
Hi HN, I’m Per, founder of Scrimba (YC S20). We’ve spent the last decade teaching people how to code with an HTML-based video format. We’ve now plugged an LLM into it, so that people can create explainer videos about anything. It’s called “Scrimba Explain”. To demo this technology for Hacker News, we built HN.watch. It’s like HN, but with explainer videos instead of articles. We create them on-the-fly the first time someone clicks on a link. While there are obvious visual drawbacks of using HTML instead of diffusion models, there are three big benefits: - Speed: Much faster to generate than pixel-based videos (just a few seconds from click to playback) - Cost: Our cost per video is ~$0.04. (Excluding image generation, which some videos utilize. Quickly blows up the cost) - Easy editing: the above benefits also make AI-assisted editing cheap & fast Our hypothesis is that if video creation goes from “dollars and minutes” to “cents and seconds”, a bunch of new use cases will be unlocked. Here are some we see already: - A video explanation of every single Pull Request (we do this internally) - Give every page in your internal/extrernal docs a video - Turn a complex article into a video in ~4 seconds (via our Chrome extension) - Course creators can quickly draft lessons before recording the real thing - People also create a lot of personal stuff stories for their kids, wedding invitations, birthdays, etc The stack is based on an open-source programming language (Imba) created by our CTO, Sindre Aarsæther. It compiles to JavaScript, so it interoperates fully with the npm + node ecosystem. You can learn more here: https://imba.io/ We’ve also built our own sync engine (OP), and a context management system for agents (Q). We feared this would make the LLMs struggle when writing code for us, as neither is in their training data (there’s very little Imba in there too). However, we’ve been pleasantly surprised to see that LLMs actually are really good at our stack. This is probably because the stack is extremely dense. Imba is compact, and so is OP, where a single declaration sets storage, sync, permissions, UI, and what the AI sees. This means there’s no translations between frontend, API, db and JSON where the model can get confused and get things wrong. Simply said, instead of using React.js, Express, Supabase, and LangChain, we built it all from scratch. Definitely suffering from the “not invented here” syndrome, lol! As for the models, we use Gemini, GPTs, Inworld, ElevenLabs, and a few others. If you want to try it out, just take your pick: - The Web UI (scrimba.com/explain) - MCP (add it to your coding agent) - ChatGPT Plugin - Chrome Extension You can find a link to all of the above in our docs: https://ift.tt/e7gantS And finally, a real pixel-based video of the tool: https://www.youtube.com/watch?v=k6rbHmBxSEs Would love to hear your feedback and if anyone has ideas for other use cases. PS: I expect quite a bit of pushback from HN for this launch, given how fan of text the HN crowd is. This kind of tool is not for everyone. But there are a lot of people today who prefer videos over text, especially in the younger generations.
Bucs say QB Mayfield out at least three weeks with dislocated thumb
Buccaneers quarterback Baker Mayfield is expected to miss at least three weeks with the dislocated thumb on his throwing hand he suffered Sunday, coach Todd Bowles announced Monday.
from www.espn.com - TOP https://ift.tt/i6xIZVf
from www.espn.com - TOP https://ift.tt/i6xIZVf
Minggu, 27 September 2026
NFL draft risers, notes off Saturday's college games: Is Colin Simmons the class' clear top edge rusher?
Will Colin Simmons be the runaway top edge for 2027? How early could Jadan Baugh get picked in April? Here's NFL draft spin off Week 4 of the CFB season.
from www.espn.com - TOP https://ift.tt/sPfSYwa
from www.espn.com - TOP https://ift.tt/sPfSYwa
Sabtu, 26 September 2026
Grades for every NBA offseason signing: Why Warriors get a B for extending Steph Curry
We're grading the biggest free agent signings and extensions, including Steph's new two-year deal.
from www.espn.com - TOP https://ift.tt/vmWpSd8
from www.espn.com - TOP https://ift.tt/vmWpSd8
Jumat, 25 September 2026
Toyota is the favorite to win the NASCAR Cup Series title, but its drivers don't feel that pressure
"I think it's Toyota's championship to lose," Bubba Wallace said. "Based on how many races we've won, it would be a shame if we didn't seal the deal."
from www.espn.com - TOP https://ift.tt/CZBHK8d
from www.espn.com - TOP https://ift.tt/CZBHK8d
Kamis, 24 September 2026
Tyson Fury vs. Anthony Joshua: What you need to know
Everything you need to know about Tyson Fury's fight with Anthony Joshua.
from www.espn.com - TOP https://ift.tt/2vFcoGQ
from www.espn.com - TOP https://ift.tt/2vFcoGQ
Rabu, 23 September 2026
Expert picks, players under pressure and rookies to watch
Which players are under the most pressure? Will any rookies step up? Who will hoist the trophy Sunday? We break down the Presidents Cup, which gets started Thursday at Medinah.
from www.espn.com - TOP https://ift.tt/IcbxjTf
from www.espn.com - TOP https://ift.tt/IcbxjTf
Selasa, 22 September 2026
Rams' McVay unsure about Nacua's status for Week 3 vs. Broncos
Sean McVay said he doesn't know wide receiver Puka Nacua's game status for Week 3 against the Broncos, but the Rams' head coach said he's hopeful Nacua will play.
from www.espn.com - TOP https://ift.tt/9zQ1jBC
from www.espn.com - TOP https://ift.tt/9zQ1jBC
Senin, 21 September 2026
MLS: Union skyrocket as unbeaten streak reaches 11 games
Can anyone stop the Philadelphia Union? Their latest win powered them up the charts in this week's edition of ESPN's MLS Power Rankings.
from www.espn.com - TOP https://ift.tt/R8Eu2on
from www.espn.com - TOP https://ift.tt/R8Eu2on
Minggu, 20 September 2026
Longtime exec Accorsi, architect of Giants' historic SB team, dies at 84
Longtime NFL executive Ernie Accorsi, the architect of the Giants team that pulled off one of the biggest upsets in Super Bowl history, has died. He was 84.
from www.espn.com - TOP https://ift.tt/Sfo0Ntc
from www.espn.com - TOP https://ift.tt/Sfo0Ntc
Sabtu, 19 September 2026
Arsenal's aura of champions evaporates after Brighton thrash them 3-0
The strengths that won Arsenal the title deserted them at Brighton, which celebrated their 125th anniversary with a thumping victory over the champions.
from www.espn.com - TOP https://ift.tt/ZUa9VB2
from www.espn.com - TOP https://ift.tt/ZUa9VB2
De Zerbi won't talk to Spurs players after loss
Roberto De Zerbi called on the "leaders" in the Tottenham squad to "lead the team" as he criticized the manner in which they conceded the first two of Aston Villa's goals.
from www.espn.com - TOP https://ift.tt/n8js5US
from www.espn.com - TOP https://ift.tt/n8js5US
Jumat, 18 September 2026
Mired in historic interception drought, Jets rule out Fitzpatrick
The Jets, mired in the longest interception drought in NFL history, will be without their most accomplished interceptor Sunday against the Packers as former All-Pro safety Minkah Fitzpatrick was ruled out with a groin injury.
from www.espn.com - TOP https://ift.tt/vK3VJhn
from www.espn.com - TOP https://ift.tt/vK3VJhn
Kamis, 17 September 2026
New top story on Hacker News: Show HN: AutoBot – live voice control for long-running AI work
Show HN: AutoBot – live voice control for long-running AI work
6 by demeyer1 | 0 comments on Hacker News.
I wanted to manage long horizon agentic workstreams via voice, then put my phone down, and have a harness manage completion - extending into full computer use. I was trying to build a personal Jarvis, so I benchmarked AutoBot to see how close I could get: - OSWorld: 32.41% (moved Sol Max from 4th to 1st, beating Opus 5) - AssistantBench: 50.70% Hermes and OpenClaw, but without needing a weekend and VMs. And with the ability to manage deep personalization AND keep strict privacy rules, storing data on encrypted disk. AutoBot is a passion project that grew out of trying to make this all work for myself. I’m sharing it here because I suspect other here have the same frustration. And I’m curious what else everyone is doing for this. It’s an MIT-licensed harness that lives within a project. Native voice lets me discuss tasks, check progress, and steer work; a local ledger tracks unfinished outputs and the evidence needed to call them done. Memory drives more autonomy over time, and “defrags and locks in learning nightly” while a heartbeat system persists execution. I’m not selling anything. If you’re building something similar for yourself, I’d love to compare notes. GitHub: https://ift.tt/VsutbUr
6 by demeyer1 | 0 comments on Hacker News.
I wanted to manage long horizon agentic workstreams via voice, then put my phone down, and have a harness manage completion - extending into full computer use. I was trying to build a personal Jarvis, so I benchmarked AutoBot to see how close I could get: - OSWorld: 32.41% (moved Sol Max from 4th to 1st, beating Opus 5) - AssistantBench: 50.70% Hermes and OpenClaw, but without needing a weekend and VMs. And with the ability to manage deep personalization AND keep strict privacy rules, storing data on encrypted disk. AutoBot is a passion project that grew out of trying to make this all work for myself. I’m sharing it here because I suspect other here have the same frustration. And I’m curious what else everyone is doing for this. It’s an MIT-licensed harness that lives within a project. Native voice lets me discuss tasks, check progress, and steer work; a local ledger tracks unfinished outputs and the evidence needed to call them done. Memory drives more autonomy over time, and “defrags and locks in learning nightly” while a heartbeat system persists execution. I’m not selling anything. If you’re building something similar for yourself, I’d love to compare notes. GitHub: https://ift.tt/VsutbUr
New top story on Hacker News: Show HN: Aclif – Agent CLI framework: one grammar, canonical names across SaaS
Show HN: Aclif – Agent CLI framework: one grammar, canonical names across SaaS
10 by chris_marino | 4 comments on Hacker News.
Simple discovery of the provider's default schemas is possible without any credentials (since they are built into the binary). If you want it to fetch customizations of your specific instance, give it credentials to your org. The repo has more details. https://ift.tt/18WL4CN
10 by chris_marino | 4 comments on Hacker News.
Simple discovery of the provider's default schemas is possible without any credentials (since they are built into the binary). If you want it to fetch customizations of your specific instance, give it credentials to your org. The repo has more details. https://ift.tt/18WL4CN
I raced against Max Verstappen ... for 10 seconds
ESPN's Nate Saunders might not have lasted long in Max vs 100, but it was still a memorable event. Here, he reflects on his karting experience with Formula 1's Max Verstappen
from www.espn.com - TOP https://ift.tt/mx0UF9S
from www.espn.com - TOP https://ift.tt/mx0UF9S
Rabu, 16 September 2026
Browns QB Watson: Focus on winning, not job security
Browns quarterback Deshaun Watson said Wednesday that his focus is on getting a win in Week 2, not his job security or the potential fan reception awaiting him in the team's home opener in Week 3.
from www.espn.com - TOP https://ift.tt/YdxNAnH
from www.espn.com - TOP https://ift.tt/YdxNAnH
Live updates, commentary as Verstappen faces go-karters in epic Silverstone challenge
Follow live updates of Formula 1 star Max Verstappen take on 100 drivers starting from the back of the grid.
from www.espn.com - TOP https://ift.tt/NWJy5fn
from www.espn.com - TOP https://ift.tt/NWJy5fn
Selasa, 15 September 2026
Cunningham reveals she talked with Silver about next WNBA commissioner
Sophie Cunningham has discussed the soon-to-be-vacant WNBA commissioner's job with Adam Silver after the NBA commissioner reached out to her.
from www.espn.com - TOP https://ift.tt/zxVsngT
from www.espn.com - TOP https://ift.tt/zxVsngT
Senin, 14 September 2026
Waiver wire: Shough, Black, Coker highlight top picks for Week 2
Eric Karabell lists who he thinks are the top available players worth adding for the week ahead.
from www.espn.com - TOP https://ift.tt/0bx51ok
from www.espn.com - TOP https://ift.tt/0bx51ok
Is Trevor Lawrence's 4-TD day a new normal? Why Jaguars' hot start faces early test
Jacksonville's QB looked dominant vs. the Browns in Week 1. Can he shake his career September woes as tough matchups loom?
from www.espn.com - TOP https://ift.tt/KqWgC0v
from www.espn.com - TOP https://ift.tt/KqWgC0v
Minggu, 13 September 2026
NFL draft risers, notes off Saturday's college games: Questions at the top of the QB class?
Should we be worried about Dante Moore's play? What has been most impressive about Jeremiah Smith? Here's NFL draft spin off Week 2 of the CFB season.
from www.espn.com - TOP https://ift.tt/CzrQV9Z
from www.espn.com - TOP https://ift.tt/CzrQV9Z
Sabtu, 12 September 2026
Week 2 live: See the best from OU-Michigan and more
Week 2 is bookended by two great helmet games: Ohio State-Texas and Oklahoma-Michigan. Here's the best of the day.
from www.espn.com - TOP https://ift.tt/oKEajZt
from www.espn.com - TOP https://ift.tt/oKEajZt
In-form Hull show why Chelsea fans are in for a rough ride this season
Chelsea have a star-studded attack which has saved them many times already this season, but their defense was exposed again vs. an impressive Hull City.
from www.espn.com - TOP https://ift.tt/Ch7gDeV
from www.espn.com - TOP https://ift.tt/Ch7gDeV
Jumat, 11 September 2026
Source: Cardinals' Love expected to play Sunday
Cardinals rookie running back, after battling a high ankle sprain since mid-August, is expected to play Sunday against the Chargers, a source told ESPN.
from www.espn.com - TOP https://ift.tt/38cR61P
from www.espn.com - TOP https://ift.tt/38cR61P
Jags' Hunter to play both ways, but plan is fluid
Travis Hunter will play both ways for the Jaguars, but how much is still fluid.
from www.espn.com - TOP https://ift.tt/m2AY4EH
from www.espn.com - TOP https://ift.tt/m2AY4EH
Kamis, 10 September 2026
New top story on Hacker News: Show HN: DOOM in the kernel, or fibers in eBPF
Show HN: DOOM in the kernel, or fibers in eBPF
6 by ayles | 2 comments on Hacker News.
Some time ago my coworker who was working on first version of Perforator ( https://ift.tt/B7Q8I5J ) kept talking about eBPF, so I got curious about its ISA and restrictions. That was around the time I read about DOOM on pregnancy test, so I thought: "what if someone run DOOM inside of Linux kernel in eBPF? Surely, with some restrictions, this should still be possible?" I started to play around with it somewhere in 2024 - first stripping DOOM to bare minimum that is needed for showcase, simplifying code and trying to pass first restrictions that I encountered - function limit, recursion, memory access and checks. I tried to do this by hand, but it was, well, tedious labor. So I changed direction and instead started to do some of the things with LLVM-passes - rewriting memory access, simplifying and rewriting loops etc. But there were just so many corners. And given absence of time and all, I forgot about this project. But couple months ago I thought: "LLMs are quite good nowadays, so why not try again this time with more hands". I tried different approaches on how memory access could be "virtualized", how loops can be made bounded, and in general - how to run unbounded logic on such a machine. When I got DOOM running, I got carried away. Approach was so "generic", so it would be a crime not to try to run more things. And now we have it - lua running on xdp hot path, llama2 working with softfloats, and even cpython, that did not fit initially into 1M verifier budget, is running there now thanks to freplace. There is ton of work to do in order to make programs run faster, to make integration easier, but working examples are already there.
6 by ayles | 2 comments on Hacker News.
Some time ago my coworker who was working on first version of Perforator ( https://ift.tt/B7Q8I5J ) kept talking about eBPF, so I got curious about its ISA and restrictions. That was around the time I read about DOOM on pregnancy test, so I thought: "what if someone run DOOM inside of Linux kernel in eBPF? Surely, with some restrictions, this should still be possible?" I started to play around with it somewhere in 2024 - first stripping DOOM to bare minimum that is needed for showcase, simplifying code and trying to pass first restrictions that I encountered - function limit, recursion, memory access and checks. I tried to do this by hand, but it was, well, tedious labor. So I changed direction and instead started to do some of the things with LLVM-passes - rewriting memory access, simplifying and rewriting loops etc. But there were just so many corners. And given absence of time and all, I forgot about this project. But couple months ago I thought: "LLMs are quite good nowadays, so why not try again this time with more hands". I tried different approaches on how memory access could be "virtualized", how loops can be made bounded, and in general - how to run unbounded logic on such a machine. When I got DOOM running, I got carried away. Approach was so "generic", so it would be a crime not to try to run more things. And now we have it - lua running on xdp hot path, llama2 working with softfloats, and even cpython, that did not fit initially into 1M verifier budget, is running there now thanks to freplace. There is ton of work to do in order to make programs run faster, to make integration easier, but working examples are already there.
Packers RB Jacobs pleads no contest to misdemeanor charges
Packers running back Josh Jacobs pleaded no contest to misdemeanor charges of battery and criminal damage to property at a court hearing in Green Bay, Wisconsin on Thursday.
from www.espn.com - TOP https://ift.tt/bv3DBya
from www.espn.com - TOP https://ift.tt/bv3DBya
Rabu, 09 September 2026
Facts vs. Feelings: Seven players who will get off to fast (or slow) starts in Week 1
What to expect from Jaxson Dart, D'Andre Swift, Quinshon Judkins, Kenneth Gainwell, Tetairoa McMIllan, Rashee Rice and Juwan Johnson in Week 1.
from www.espn.com - TOP https://ift.tt/SomRM6X
from www.espn.com - TOP https://ift.tt/SomRM6X
Selasa, 08 September 2026
2026 Ballon d'Or: Full men's, women's shortlists, host city change, more
The 2026 Ballon d'Or shortlists are out. Here's who could win the biggest individual prizes in men's and women's soccer.
from www.espn.com - TOP https://ift.tt/ck5TDv1
from www.espn.com - TOP https://ift.tt/ck5TDv1
Senin, 07 September 2026
Arsenal show Chelsea, Premier League who's boss; Inter the team to beat in Serie A
Arsenal looked rattled against Chelsea thanks to an early goal, but Mikel Arteta's side quickly showed why it's the team to beat in England. Plus Inter prove they are Serie A's top team, as Gab Marcotti recaps the weekend.
from www.espn.com - TOP https://ift.tt/MWX8nGD
from www.espn.com - TOP https://ift.tt/MWX8nGD
Minggu, 06 September 2026
Another game, more dropped points: Are Man United still a work in progress?
Michael Carrick had his head in his hands: Manchester United are just three games into the Premier League season, but they've already dropped points twice.
from www.espn.com - TOP https://ift.tt/TfJ8ZkA
from www.espn.com - TOP https://ift.tt/TfJ8ZkA
New top story on Hacker News: We monitor internal coding agents for misalignment
We monitor internal coding agents for misalignment
27 by lukaspetersson | 14 comments on Hacker News.
27 by lukaspetersson | 14 comments on Hacker News.
Sabtu, 05 September 2026
Tressel enshrined in Ohio State's ring of honor
Jim Tressel, who coached Ohio State to the 2002 National Championship, was enshrined into the Buckeyes' ring of honor on Saturday.
from www.espn.com - TOP https://ift.tt/Zkvuadz
from www.espn.com - TOP https://ift.tt/Zkvuadz
Katie Taylor vs. Flora Pili: Latest updates, results, analysis from Croke Park
ESPN has live updates as Katie Taylor makes her final ring-walk, taking on Flora Pili for the undisputed junior welterweight titles in Dublin.
from www.espn.com - TOP https://ift.tt/5HRsrJh
from www.espn.com - TOP https://ift.tt/5HRsrJh
Mali shocks Spain at FIBA Women's World Cup
Mali pulled off the first stunner of the Women's Basketball World Cup by shocking Spain 82-73.
from www.espn.com - TOP https://ift.tt/ZmRaxfG
from www.espn.com - TOP https://ift.tt/ZmRaxfG
Jumat, 04 September 2026
New top story on Hacker News: Show HN: TERMy – A fast terminal assistant that does not use LLMs
Show HN: TERMy – A fast terminal assistant that does not use LLMs
1 by gioscarab | 0 comments on Hacker News.
I love research and development, you may have heard of me because of PJON (Padded Jittering Operative Network). It is a network protocol I started developing in 2010, which was recently implemented in silicon by the ETH Zurich university thanks to the research of Pius Sieber. I am excited to share with you TERMy, a terminal assistant built on top of the NPC-Forge framework. Unlike everything else being built today, TERMy does not use embeddings, machine-learning or LLMs. It runs on the CPU (even on a Raspberry Pi Zero) both in the terminal or client-side in a browser tab and responds in milliseconds. It is a cynical but very knowledgeable Linux terminal assistant that translates your natural language into shell commands without relying on a single artificial neuron. I had a chance to focus for 2 months on my personal projects since early July, during the strange times of AI price hikes and the end of subsidized tokenmaxing. I was curious to see if I could develop from scratch a terminal assistant capable of handling simple natural language requests. I have a bad memory and got used to ask to copilot "activate the virtual environment" or similar trivial operations spending a non negligible sum every month. I started thinking, maybe I can do something to make my workflow more efficient? Do I really need trillions of parameters to accomplish those tasks? How it Works When you type a prompt, it goes through a lightweight NLU pipeline written in ~1000 lines of Python that implement the following steps: 1. Strip expletives, interjections, encouraging, discouraging and thanking words (remove noise) 2. Sentiment analysis 3. Exact Match (very fast) 4. Template Match (slower) 5. Probabilistic Match (even slower) Step 5 relies on: 1. IDF (Inverse Document Frequency) to identify rare words. 2. BOW (Bag Of Words) to accommodate word inversions. 3. IDF weighted Levenshtein to safely handle typos. Permission gating is hardcoded into the dataset and enforced for all potentially destructive commands, so it's inherently safer than letting an unpredictable LLM run wild on your machine. - TERMy in operation: https://www.youtube.com/watch?v=qeIp0xePLBg - Variance and typo tolerance: https://www.youtube.com/watch?v=tQvGDk6fkk0 - Copilot integration: https://www.youtube.com/watch?v=Wzzouhq2a8A - Advanced features: https://www.youtube.com/watch?v=qeIp0xePLBg - Source Code: https://ift.tt/bMvQfV3
1 by gioscarab | 0 comments on Hacker News.
I love research and development, you may have heard of me because of PJON (Padded Jittering Operative Network). It is a network protocol I started developing in 2010, which was recently implemented in silicon by the ETH Zurich university thanks to the research of Pius Sieber. I am excited to share with you TERMy, a terminal assistant built on top of the NPC-Forge framework. Unlike everything else being built today, TERMy does not use embeddings, machine-learning or LLMs. It runs on the CPU (even on a Raspberry Pi Zero) both in the terminal or client-side in a browser tab and responds in milliseconds. It is a cynical but very knowledgeable Linux terminal assistant that translates your natural language into shell commands without relying on a single artificial neuron. I had a chance to focus for 2 months on my personal projects since early July, during the strange times of AI price hikes and the end of subsidized tokenmaxing. I was curious to see if I could develop from scratch a terminal assistant capable of handling simple natural language requests. I have a bad memory and got used to ask to copilot "activate the virtual environment" or similar trivial operations spending a non negligible sum every month. I started thinking, maybe I can do something to make my workflow more efficient? Do I really need trillions of parameters to accomplish those tasks? How it Works When you type a prompt, it goes through a lightweight NLU pipeline written in ~1000 lines of Python that implement the following steps: 1. Strip expletives, interjections, encouraging, discouraging and thanking words (remove noise) 2. Sentiment analysis 3. Exact Match (very fast) 4. Template Match (slower) 5. Probabilistic Match (even slower) Step 5 relies on: 1. IDF (Inverse Document Frequency) to identify rare words. 2. BOW (Bag Of Words) to accommodate word inversions. 3. IDF weighted Levenshtein to safely handle typos. Permission gating is hardcoded into the dataset and enforced for all potentially destructive commands, so it's inherently safer than letting an unpredictable LLM run wild on your machine. - TERMy in operation: https://www.youtube.com/watch?v=qeIp0xePLBg - Variance and typo tolerance: https://www.youtube.com/watch?v=tQvGDk6fkk0 - Copilot integration: https://www.youtube.com/watch?v=Wzzouhq2a8A - Advanced features: https://www.youtube.com/watch?v=qeIp0xePLBg - Source Code: https://ift.tt/bMvQfV3
Kamis, 03 September 2026
Pirates rookie Griffin eyes Friday return from IL
Pirates rookie Konnor Griffin said he feels good and is hopeful of being activated off the 60-day IL on Friday when Pittsburgh opens a 3-game series against the Angels.
from www.espn.com - TOP https://ift.tt/ZpkIrmz
from www.espn.com - TOP https://ift.tt/ZpkIrmz
Rabu, 02 September 2026
Monza personifies the passion that enticed Leclerc to commit his future to Ferrari
Ferrari hasn't been a title contender in some time, but Charles Leclerc is confident that will change. Until then, he's enjoying being an icon in Italy.
from www.espn.com - TOP https://ift.tt/VITQwnK
from www.espn.com - TOP https://ift.tt/VITQwnK
Selasa, 01 September 2026
Sources: Chelsea's Enzo set for $169M City move
Manchester City have agreed a Premier League record-equalling £125 million ($169m) deal with Chelsea for Enzo Fernández, sources have told ESPN.
from www.espn.com - TOP https://ift.tt/kdpGSFn
from www.espn.com - TOP https://ift.tt/kdpGSFn
Langganan:
Postingan (Atom)