developer

ample: My nginx configuration is giving me [SPECIFIC ERROR] for [CONFIG FILE]. What's wrong and how can I fix it? The domains where it does not work are: 1) You have some generic code with domain/company/whatever specific edge cases. The edge cases, broadly speaking, no matter how well documented, will not be handled well by the model. Edge cases are exactly that; edge cases; the common medium of 'how to x' does not cover edge cases; the edge cases will not be covered and the results will require you to review and complete them manually. 2) You have some specific piece of code you want to refactor 'to solve xxx', but the code is not covered well by tests. LLMs struggle to refactor existing code, and the difficulty is proportional to the code length. There are technical reasons for this (mainly randomizing token weights), but tldr; it's basically a crap shot. Might work. Might not. If you have no tests who knows? You have to manually verify both the new functionality and the old functionality, but maybe it helps a bit, at scale, for trivial problems. 3) You're doing something obscure or using a new library / new version of the library. The LLM will have no context for this, and will generate rubbish / old deprecated content. Obscure requirements have an unfortunate tendency to mimic the few training examples that exist, and may generate verbatim copies, depending on the model you use. ... So. Concrete advice: 1) sigh~ > a friend of mine came and suggested that I use Retrieval-Augmented Generation (RAG), I have yet to try it, with a setup Langchain + Ollama. Ignore this advice. RAG and langchain are not the solutions you are looking for. 2) Use a normal coding assistant like copilot. This is the most effective way to use AI right now. There are some frameworks that let you use open source models if you don't want to use openAI. 3) Do not attempt to bulk generate code. AI coding isn't at that level. Right now, the tooling is primitive, and large scale coherent code generation is... not impossible, but it is difficult (see below). You will be more effective using an existing proven path that uses 'copilot' style helpers. However... ...if you do want to pursue code generation, here's a broad blueprint to follow: - decompose your task into steps - decompose you steps in functions - generate or write tests and function definitions - generate an api specification (eg. .d.ts file) for your function definitions - for each function definition, generate the code for the function passing the api specification in as the context. eg. "Given functions x, y, z with the specs... ; generate an implementation of q that does ...". - repeated generate multiple outputs for the above until you get one that passes the tests you wrote. This approach broadly scales to reasonably complex problems, so long as you partition your problem into module sized chunks. I personally like to put something like "you're building a library/package to do xxx" or "as a one file header" as a top level in the prompt, as it seems to link into the 'this should be isolated and a package' style of output. However, I will caveat this with two points: 1) You generate a lot of code this way, and that's expensive if you use a charge-per-completion API. 2) The results are not always coherent and functions tend to (depending on the model, eg. 7B mistral) inline implementations for 'trivial' functions instead of using functions (eg. if you define Vector::add, the model will 50/50 just go a = new Vector(a.x + b.x, a.y + b.y)). I've found that the current models other than GPT4 are prone to incoherence as the problem size scales. 7B models, specifically, perform significantly worse than larger models.

shallow understanding of the code base despite being able to request more. But hopefully this gets better as LLMs improve or I get better at prompt engineering. [0] https://github.com/smol-ai/developer

es. 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 - Conversationally iterate on code edits over selected regions which are streamed directly into your editor window - 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') - real-time automatic docstring generation triggered by saving a file or making a git commit - run tools like typecheckers and static analyzers in the background and implement fixes based on their output - 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 — 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.

d files or ask for followup based on your response. It does stuff like make sure the file exists, show you a git diff of changes, respond if the LLM is not following the system prompt, etc. It also addresses the problem that a lot of these tools eat up way too many tokens so a single prompt to something like smol-dev would eat up a few dollars on every iterations. It's still very much a work in progress and i'll prob do a show hn next week but I would love some feedback [0] https://github.com/breeko/j-dev [1] https://github.com/smol-ai/developer [2] https://github.com/breeko/j-dev/blob/master/src/prompts/syst...

ing at, say, https://github.com/minimaxir/simpleaichat/tree/main/simpleai... or https://github.com/smol-ai/developer and digging in to the prompts, what goes in system vs assistant roles, how you guide the LLM, etc.

ontext as well. I broke out smol-developer's prompts and modified them to include "ask any clarifying questions you need to build the [list of files/dependencies/source]" and then was able to provide some more feedback during the development process. I also started off with that trick of "You are an AI prompt engineer, gather requirements and ask any follow up questions that are necessary to generate the perfect prompt, optimized to feed into GPT, to achieve the user's goals." trick, did some back and forth during which it asked some fascinating questions, including at least one that wasn't even on my radar. I then used that generated prompt in the following steps with smol-developer.