dwata

I have a project on very similar lines, https://github.com/brainless/dwata, which I have not been developing for the past few months. I have been meaning to get back to it and I really like what I see on your project page. My aim is to build a truly local app using only tiny/small models. I have had really good results from Qwen 3.5 1B, 4B, etc. Also, Gliner or similar models for different uses. SQLite + sqlite-vec + Tantivy + a tiny embedding model will stay as my go to. In my case, coding agent is a separate product. I have https://github.com/brainless/nocodo for that. nocodo is also built for tiny/small models from the ground up. And recently I started building a wGPU based UI framework to build both these apps as native UI apps in Rust: https://github.com/brainless/akar. I also want e2e encrypted team/family sharing in my products. Th

I am interested in MetalRT. I am an indie builder, focused mostly on building products with LLM assistance that run locally. Like: https://github.com/brainless/dwata I would be interested if MetalRT can be used by other products, if you have some plans for open source products?

I wanted to have all my emails (and files) scanned for financial data. Transactions, Bills (I may not have paid). I wanted this to run entirely locally and not depend on a Large Language Model from a cloud provider. I initially started with Google Gemini 3 Flash but I switched to Ollama + Ministral 3:3b. The extraction is not exhaustive and there is much to improve but this is working. dwata runs locally, runs a web backend and the gui runs in browser. Connects to emails, downloads them. Then we can run the financial template detection. It checks for similar looking emails, grouped by sender. Then sends a sample from each cluster to LLM agent. The LLM is asked to find out the parts of text that look like the data we are looking for. dwata then searches for the variables/values that LLM gave in the email, creates a template by replacing th

Local models, particularly the new ones would be really useful in many situations. They are not for general chat but if tools use them in specific agents, the results are awesome. I built https://github.com/brainless/dwata to submit for Google Gemini Hackathon, and focused on an agent that would replace email content with regex to extract financial data. I used Gemini 3 Flash. After submitting to the contest, I kept working on branch: reverse-template-based-financial-data-extraction to use Ministral 3:3b. I moved away from regex detection to a reverse template generation. Like Jinja2 syntax but in reverse, from the source email. Financial data extraction now works OK ish and I am constantly improving this to aim for a launch soon. I will try with Qwen 3.5 Small, maybe 4b model. Both Ministral 3:3b and Qwen 3.5 Small:4b will fit on the sma

The trick with small models is what you ask them to do. I am working on a data extraction app (from emails and files) that works entirely local. I applied for Taalas API because it would be awesome fit. dwata: Entirely Local Financial Data Extraction from Emails Using Ministral 3 3B with Ollama: https://youtu.be/LVT-jYlvM18 https://github.com/brainless/dwata

I know it is not easy to see the benefits of small models easily but this is what I am building for (1). I created a product for Google Gemini 3 Hackathon and I used Gemini 3 Flash (2). I tested locally using Ministral 3B and it was promising. Definitely will need work. But 8B/14B may give awesome results. I am building a data extraction software on top of emails, attachments, cloud/local files. I use a reverse template generation with only variable translation done by LLMs (3). Small models are awesome for this (4). I just applied for API access. If privacy policies are a fit, I would love to enable this for MVP launch. 1. https://github.com/brainless/dwata 2. https://youtu.be/Uhs6SK4rocU 3. https://github.com/brainless/dwata/tree/feature/reverse-temp... 4. https://github.com/brainless/dwata/tree/feature/reverse-temp...

For the last couple weeks I have been building dwata and I am going to submit today for Google Gemini Hackathon. https://github.com/brainless/dwata dwata is built on the idea of multiple, task-specific agents. Right now it has only one agent that can be run on an email to extract regex patterns for financial data. This enables high performance data extraction from emails or documents (in future) without sending each email to an LLM. dwata has an email scan which tests simple keywords and regex patterns, groups by sender emails, sorts by number of emails per sender (highest first), and filters out groups where the emails do not seem to be from a template (typical transaction emails are from templates). This is deterministic code in Rust. Then dwata can use the regex builder AI agent to take one email from the group and build a regex patter

As a solo founder trying again (1) after many failed attempts, I can only say that my investment in Rust is paying off. A lot. I have worked with PHP in my early days (2006...), for about 6 years. I wanted to move to a language more structured. When I switched to Python I felt it was worth my investment. Working as a web developer, JavaScript was always there and I could not change that. After working with Python for 11 years, I felt if I attempt to be a founder again, I needed two things from a language - structure in developing my ideas, peace of mind to take a product to users' hands. I was OK to learn syntax, but I can not become a legendary developer with magical knowledge of some language. Refactoring the codebase is wildly different since I get so much support from the compiler/type system. And, if it compiles, it will simply not h

I have been working full-time since early 2024 on a locally running app, where I stick to open file formats and SQLite for storage. I would like to make this "local-first" - I feel I'm already on the way. It is an open source app to download pervade data and provide fast search, offline access, automatic labels, categories, summaries, etc Starting with emails, calendars and attachments. Integrations for Slack, LinkedIn, Stripe, databases, even web crawl are coming... I plan to have real-time collaboration on top to make money (ideally with end to end encryption). I take privacy very seriously. I do not even have OAuth2 apps for Google to access your data - you have to create those yourself (there is documentation in the app). This branch has updated readme: https://github.com/brainless/dwata/tree/feature/prepare_mvp_...

Open source app to download emails, provide fast search, offline access, automatic labels, categories, summaries, surface up what needs our attention, etc. (some features need AI). Starting with emails, calendars and attachments. Integrations for Slack, LinkedIn, Stripe, databases, even web crawl are coming... This branch has updated readme: https://github.com/brainless/dwata/tree/feature/prepare_mvp_...

I use RustRover/VS Code + Codeium or Zed + Supermaven and I have used Copilot before. To be honest it takes some time to get used to the flow. I have turned them off multiple times and finally the workflow has set well in my brain. Now I feel very productive with them. I work full-time for my own product (very early stage) but I am happy to share my own journey of using AI code assistants. Please feel free to check the commits: https://github.com/brainless/dwata

I have been looking at your extension for the last couple weeks and I think I will end up using it. I am also looking at running qdrant locally. I am creating an open source AI studio (desktop app) (1). It is in very early stages but it is fun to get to know this landscape. I would be happy to contribute to your project in any way I can. And thank you for this project. 1. https://github.com/brainless/dwata

I love this project. This is part of my long term goal for my main product [1] which is all about making databases easier to work with for everyone while abstracting away SQL. The use of ML/RL is exactly the method I have in mind for long term. As I am not from that background, so my immediate goals are more like a set of best practices and migration management with a Python layer (SQLAlchemy, sqlalchemy-migrate and automation on top). [1] https://github.com/brainless/dwata

I would be happy to talk if I can help. I am building https://github.com/brainless/dwata but from a different angle. It is not a builder as such and has no SQL in the interface at all. It is where non-technical folks can do everything - dig into data, merge data from CSV even, create reports. SQL is generated automatically from schema reflection. There is no engineering help needed, at least at this stage and it does multi-table JOINs by itself already.