As part of Rift, we're releasing an open-source VSCode extension which shows what is possible today with a language server that can mediate between LLMs and IDEs, and which takes a first step towards a future where you can build and iterate on entire projects with only natural language gestures.
Today, with Rift, you can do all of the following without ever leaving VSCode or copy-pasting from a browser chat window:
- Generate an entire workspace, or a module based on other parts of your codebase\n- Conversationally iterate on code edits over selected regions which are streamed directly into your editor window\n- Request, review, and iterate on codebase-wide, multi-file edits
Everything runs as an agent through Rift's Agents API. You can define your own agents that can interact directly with your IDE and codebase through the language server, and can run arbitrary Python code that can do things like:
- provide a conversational UI over the git CLI (e.g. 'suggest fixes for these merge conflicts, then rebase over myBranch')\n- real-time automatic docstring generation triggered by saving a file or making a git commit\n- run tools like typecheckers and static analyzers in the background and implement fixes based on their output\n- automated code review in your IDE with context from GitHub
We're proud to integrate with open-source projects like Smol Developer (https://www.github.com/smol-ai/developer), GPT Engineer (https://www.github.com/antonosika/gpt-engineer), and Aider (https://www.github.com/paul-gauthier/aider).
Language servers are ubiquitous and see every action you take in an IDE. Rift will soon support recording training data from your interactions so that you can work with a team of truly personalized AI software engineers.
At Morph Labs (https://morph.so), we believe that the future of software development should be transparent, open-source, and defined collectively by the developer community. The Rift language server and extension will continue to be open-source, extensible, and privacy/security aware with opt-in telemetry and first-class support for custom endpoints and on-device models \u2014 no closed source language server binaries, no API calls to mysterious proxy servers.
We would love for you to try out Rift and give us feedback! Please let us know what you think in the comments."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Rift \u2013 Open-source AI-native language server for your personal AI SWE"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/morph-labs/rift"}},"_tags":["story","author_jessemhan","story_37065734","show_hn"],"author":"jessemhan","children":[37066152,37066247,37066813,37067585],"created_at":"2023-08-09T17:13:44Z","created_at_i":1691601224,"num_comments":5,"objectID":"37065734","points":17,"story_id":37065734,"story_text":"Howdy y'all, I'm Jesse and I'm pleased to share Rift (https://www.github.com/morph-labs/rift), an open-source AI-native language server for the IDEs of the future. Software will soon be written mostly by AI SWEs working alongside humans. Codebases and the development environments around them will soon become living artifacts that can anticipate, maintain context on, and execute your every intention. Software development is rapidly evolving and key infrastructure like language servers must evolve with it to support a new generation of developer tools.
As part of Rift, we're releasing an open-source VSCode extension which shows what is possible today with a language server that can mediate between LLMs and IDEs, and which takes a first step towards a future where you can build and iterate on entire projects with only natural language gestures.
Today, with Rift, you can do all of the following without ever leaving VSCode or copy-pasting from a browser chat window:
- Generate an entire workspace, or a module based on other parts of your codebase\n- Conversationally iterate on code edits over selected regions which are streamed directly into your editor window\n- Request, review, and iterate on codebase-wide, multi-file edits
Everything runs as an agent through Rift's Agents API. You can define your own agents that can interact directly with your IDE and codebase through the language server, and can run arbitrary Python code that can do things like:
- provide a conversational UI over the git CLI (e.g. 'suggest fixes for these merge conflicts, then rebase over myBranch')\n- real-time automatic docstring generation triggered by saving a file or making a git commit\n- run tools like typecheckers and static analyzers in the background and implement fixes based on their output\n- automated code review in your IDE with context from GitHub
We're proud to integrate with open-source projects like Smol Developer (https://www.github.com/smol-ai/developer), GPT Engineer (https://www.github.com/antonosika/gpt-engineer), and Aider (https://www.github.com/paul-gauthier/aider).
Language servers are ubiquitous and see every action you take in an IDE. Rift will soon support recording training data from your interactions so that you can work with a team of truly personalized AI software engineers.
At Morph Labs (https://morph.so), we believe that the future of software development should be transparent, open-source, and defined collectively by the developer community. The Rift language server and extension will continue to be open-source, extensible, and privacy/security aware with opt-in telemetry and first-class support for custom endpoints and on-device models \u2014 no closed source language server binaries, no API calls to mysterious proxy servers.
We would love for you to try out Rift and give us feedback! Please let us know what you think in the comments.","title":"Show HN: Rift \u2013 Open-source AI-native language server for your personal AI SWE","updated_at":"2024-09-20T14:51:43Z","url":"https://github.com/morph-labs/rift"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"trees101"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"obligatory plug for aider:\nhttps://github.com/paul-gauthier/aider\nhttps://aider.chat/docs/usage.html
I run aider in a terminal, and separately review and manually code in VSCode. Usint the `--no-auto-commits` switch means that I can immediately view the diffs in the nice VSCode diff view and it's very easy to do hybrid manual and AI coding.\nThere are plenty of handy settings (see /help), for example you can /ask questions about your code.
`aider --no-auto-commits --cache-prompts --no-stream --cache-keepalive-pings 6 --no-suggest-shell-commands`"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Code Smarter, Not Harder: Developing with Cursor and Claude Sonnet"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://jstoppa.com/posts/artificial-intelligence/fundamentals/code-smarter-not-harder-developing-with-cursor-and-claude-sonnet/post/"}},"_tags":["comment","author_trees101","story_41429469"],"author":"trees101","comment_text":"obligatory plug for aider:\nhttps://github.com/paul-gauthier/aider\nhttps://aider.chat/docs/usage.html
I run aider in a terminal, and separately review and manually code in VSCode. Usint the `--no-auto-commits` switch means that I can immediately view the diffs in the nice VSCode diff view and it's very easy to do hybrid manual and AI coding.\nThere are plenty of handy settings (see /help), for example you can /ask questions about your code.
`aider --no-auto-commits --cache-prompts --no-stream --cache-keepalive-pings 6 --no-suggest-shell-commands`","created_at":"2024-09-03T02:28:11Z","created_at_i":1725330491,"objectID":"41430695","parent_id":41429469,"story_id":41429469,"story_title":"Code Smarter, Not Harder: Developing with Cursor and Claude Sonnet","story_url":"https://jstoppa.com/posts/artificial-intelligence/fundamentals/code-smarter-not-harder-developing-with-cursor-and-claude-sonnet/post/","updated_at":"2025-03-04T17:36:16Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"SparkyMcUnicorn"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"Sonnet 3.5 has a max output of 8192 tokens[0].
I break most tasks down into parts. Aider[1] is essential to my workflow and helps with this as well, and it's a fantastic tool to learn from. In fact, as of v0.52 I'm able to remove some of my custom code to run and test.
Started playing around with adding Nous[2] as well (aider is its code editing agent), but not enough that I'm using it practically yet.
[0] https://docs.anthropic.com/en/docs/about-claude/models
[1] https://github.com/paul-gauthier/aider/
[2] https://github.com/TrafficGuard/nous"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Anthropic publishes the 'system prompts' that make Claude tick"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://techcrunch.com/2024/08/26/anthropic-publishes-the-system-prompt-that-makes-claude-tick/"}},"_tags":["comment","author_SparkyMcUnicorn","story_41364637"],"author":"SparkyMcUnicorn","children":[41437122],"comment_text":"Sonnet 3.5 has a max output of 8192 tokens[0].
I break most tasks down into parts. Aider[1] is essential to my workflow and helps with this as well, and it's a fantastic tool to learn from. In fact, as of v0.52 I'm able to remove some of my custom code to run and test.
Started playing around with adding Nous[2] as well (aider is its code editing agent), but not enough that I'm using it practically yet.
[0] https://docs.anthropic.com/en/docs/about-claude/models
[1] https://github.com/paul-gauthier/aider/
[2] https://github.com/TrafficGuard/nous","created_at":"2024-08-29T13:29:36Z","created_at_i":1724938176,"objectID":"41390605","parent_id":41374397,"story_id":41364637,"story_title":"Anthropic publishes the 'system prompts' that make Claude tick","story_url":"https://techcrunch.com/2024/08/26/anthropic-publishes-the-system-prompt-that-makes-claude-tick/","updated_at":"2024-09-20T17:44:01Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"IanCal"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"> Thanks! "Nearby edits" mean edits in the same file or the whole workspace?
For the autocomplete, in the same file. So proposing adding more logging when you add a few statements, changing an error check, adding something to the class def or constructor.
They do have a multi-file editing thing called "composer" I think, which I used to make larger changes to an app (e.g. add a new page that lists all the X, and it creates that and the links to it in the other pages).
You might also be interested in aider https://github.com/paul-gauthier/aider for larger changes."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Avante.nvim: Use Your Neovim Like Using Cursor AI IDE"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/yetone/avante.nvim"}},"_tags":["comment","author_IanCal","story_41353835"],"author":"IanCal","children":[41356134],"comment_text":"> Thanks! "Nearby edits" mean edits in the same file or the whole workspace?
For the autocomplete, in the same file. So proposing adding more logging when you add a few statements, changing an error check, adding something to the class def or constructor.
They do have a multi-file editing thing called "composer" I think, which I used to make larger changes to an app (e.g. add a new page that lists all the X, and it creates that and the links to it in the other pages).
You might also be interested in aider https://github.com/paul-gauthier/aider for larger changes.","created_at":"2024-08-26T10:01:05Z","created_at_i":1724666465,"objectID":"41355586","parent_id":41354979,"story_id":41353835,"story_title":"Avante.nvim: Use Your Neovim Like Using Cursor AI IDE","story_url":"https://github.com/yetone/avante.nvim","updated_at":"2024-10-10T12:28:31Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"westurner"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"Gartner "Magic Quadrant for AI Code Assistants" (2024)\nhttps://www.gartner.com/doc/reprints?id=1-2IKO4MPE&ct=240819...
Additional criteria for assessing AI code assistants from https://news.ycombinator.com/item?id=40478539 re: Text-to-SQL bemchmarks :
codefuse-ai/Awesome-Code-LLM > Analysis of AI-Generated Code, Benchmarks: \nhttps://github.com/codefuse-ai/Awesome-Code-LLM :
> 8.2. Benchmarks: *\nIntegrated Benchmarks, \nProgram Synthesis, \nVisually Grounded Program Synthesis,\nCode Reasoning and QA,\nText-to-SQL, \nCode Translation, \nProgram Repair, \nCode Summarization, \nDefect/Vulnerability Detection, \nCode Retrieval, \nType Inference, \nCommit Message Generation, \nRepo-Level Coding*
OT did not assess:
Aider: \nhttps://github.com/paul-gauthier/aider :
> Aider works best with GPT-4o & Claude 3.5 Sonnet and can connect to almost any LLM.
> Aider has one of the top scores on SWE Bench. SWE Bench is a challenging software engineering benchmark where aider solved real GitHub issues from popular open source projects like django, scikitlearn, matplotlib, etc
SWE Bench benchmark: \nhttps://www.swebench.com/"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"GitHub Named a Leader in the Gartner First Magic Quadrant for AI Code Assistants"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://github.blog/news-insights/company-news/github-named-a-leader-in-the-gartner-first-ever-magic-quadrant-for-ai-code-assistants/"}},"_tags":["comment","author_westurner","story_41332294"],"author":"westurner","comment_text":"Gartner "Magic Quadrant for AI Code Assistants" (2024)\nhttps://www.gartner.com/doc/reprints?id=1-2IKO4MPE&ct=240819...
Additional criteria for assessing AI code assistants from https://news.ycombinator.com/item?id=40478539 re: Text-to-SQL bemchmarks :
codefuse-ai/Awesome-Code-LLM > Analysis of AI-Generated Code, Benchmarks: \nhttps://github.com/codefuse-ai/Awesome-Code-LLM :
> 8.2. Benchmarks: *\nIntegrated Benchmarks, \nProgram Synthesis, \nVisually Grounded Program Synthesis,\nCode Reasoning and QA,\nText-to-SQL, \nCode Translation, \nProgram Repair, \nCode Summarization, \nDefect/Vulnerability Detection, \nCode Retrieval, \nType Inference, \nCommit Message Generation, \nRepo-Level Coding*
OT did not assess:
Aider: \nhttps://github.com/paul-gauthier/aider :
> Aider works best with GPT-4o & Claude 3.5 Sonnet and can connect to almost any LLM.
> Aider has one of the top scores on SWE Bench. SWE Bench is a challenging software engineering benchmark where aider solved real GitHub issues from popular open source projects like django, scikitlearn, matplotlib, etc
SWE Bench benchmark: \nhttps://www.swebench.com/","created_at":"2024-08-23T21:25:39Z","created_at_i":1724448339,"objectID":"41333249","parent_id":41332294,"story_id":41332294,"story_title":"GitHub Named a Leader in the Gartner First Magic Quadrant for AI Code Assistants","story_url":"https://github.blog/news-insights/company-news/github-named-a-leader-in-the-gartner-first-ever-magic-quadrant-for-ai-code-assistants/","updated_at":"2024-09-20T17:38:02Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"trees101"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"`pipx install aider-chat`
(https://aider.chat/docs/install/pipx.html will install it globally on your system within its own python environment. This way you can use aider to work on any python project, even if that project has conflicting dependencies.)
I run it with these settings:\n`aider --sonnet --no-auto-commits--cache-prompts`
if you want to run the most bleeding edge version at the same time without breaking anything,
`pipx install --suffix=@dev git+https://github.com/paul-gauthier/aider.git`
then run it with\n`aider@dev`
`aider@dev` ... with your desired settings"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"I'm tired of fixing customers' AI generated code"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://medium.com/@thetateman/im-tired-of-fixing-customers-ai-generated-code-94816bde4ceb"}},"_tags":["comment","author_trees101","story_41315138"],"author":"trees101","comment_text":"`pipx install aider-chat`
(https://aider.chat/docs/install/pipx.html will install it globally on your system within its own python environment. This way you can use aider to work on any python project, even if that project has conflicting dependencies.)
I run it with these settings:\n`aider --sonnet --no-auto-commits--cache-prompts`
if you want to run the most bleeding edge version at the same time without breaking anything,
`pipx install --suffix=@dev git+https://github.com/paul-gauthier/aider.git`
then run it with\n`aider@dev`
`aider@dev` ... with your desired settings","created_at":"2024-08-22T20:06:46Z","created_at_i":1724357206,"objectID":"41323942","parent_id":41320531,"story_id":41315138,"story_title":"I'm tired of fixing customers' AI generated code","story_url":"https://medium.com/@thetateman/im-tired-of-fixing-customers-ai-generated-code-94816bde4ceb","updated_at":"2024-09-20T17:37:01Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"trees101"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["paul","gauthier","aider"],"value":"also, aider has added experimental support for prompt caching: https://github.com/paul-gauthier/aider/issues/1086#issuecomm...
I really like the control you get with aider over the LLM context. You can /add or /drop source code, markdown notes. You can /clear the chat. /tokens shows you the context and the cost, you can see what each prompt will cost you.
I find aider best used in conjunction with a git diff view in VSCode, I run aider with --no-auto-commits and then manually review each time in VSCode.
I'm keen to learn any AI coding workflows if anyone has any links. I've benefitted greatly from tips such as using type hints and documentation for the LLM's benefit."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Fine-tuning now available for GPT-4o"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://openai.com/index/gpt-4o-fine-tuning/"}},"_tags":["comment","author_trees101","story_41301673"],"author":"trees101","comment_text":"also, aider has added experimental support for prompt caching: https://github.com/paul-gauthier/aider/issues/1086#issuecomm...
I really like the control you get with aider over the LLM context. You can /add or /drop source code, markdown notes. You can /clear the chat. /tokens shows you the context and the cost, you can see what each prompt will cost you.
I find aider best used in conjunction with a git diff view in VSCode, I run aider with --no-auto-commits and then manually review each time in VSCode.
I'm keen to learn any AI coding workflows if anyone has any links. I've benefitted greatly from tips such as using type hints and documentation for the LLM's benefit.","created_at":"2024-08-20T21:27:15Z","created_at_i":1724189235,"objectID":"41304253","parent_id":41302999,"story_id":41301673,"story_title":"Fine-tuning now available for GPT-4o","story_url":"https://openai.com/index/gpt-4o-fine-tuning/","updated_at":"2024-09-20T17:34:59Z"}],"hitsPerPage":8,"nbHits":97,"nbPages":13,"page":0,"params":"query=%22paul+gauthier%22+aider&hitsPerPage=8&advancedSyntax=true&analyticsTags=backend","processingTimeMS":8,"processingTimingsMS":{"_request":{"roundTrip":20},"afterFetch":{"merge":{"mergeLoop":{"prepareNextHit":3,"total":3},"total":4},"total":4},"fetch":{"query":2,"total":3},"total":8},"query":"\"paul gauthier\" aider","serverTimeMS":9}