The graveyard of good ideas
Solutions that were tried, and what stopped them.
Patents that were rejected or abandoned, products that shut down, projects that went quiet. Borrow the idea, avoid the dead end.
1,900 solutions
- startupShut down2024
Launch HN: Patchwork (YC S24) – Automatically add structured logs to your code
Hey HN! We are Sam, Alex and Ben from Patchwork Technologies ( https://getpatchwork.io ). Patchwork automatically adds actionable structured logs to your code. Big picture: we are building a next generation logging and observability platform which gives engineers the rich debugging context they need. There’s a demo video ( https://youtu.be/ObIepiXfVx0 ), as well as a demo^ instance with some results for you to explore at https://hackernews.getpatchwork.io (no sign-in required!) The demo shows the analysis and improvement (against a style guide) of existing logs. It’s set up for two OSS repositories, https://github.com/elastic/cloud-on-k8s and our awesome friends at Glasskube ( https://github.com/glasskube/glasskube ). We welcome suggestions of other OSS Go repositories that you would like to see added! Why we are
Why it failed: Hey HN! We are Sam, Alex and Ben from Patchwork Technologies ( https://getpatchwork.io ). Patchwork automatically adds actionable structured logs to your code. Big picture: we are building a next generation logging and observability platform which gives engineers the ri
benjaminfh
- startupShut down2024
Show HN: Peerdb Streams – Simple, native Postgres change data capture
Hello HN, I am Sai Srirampur, one of the Co-founders of PeerDB. ( https://github.com/PeerDB-io/peerdb ). We spent the past 7 months building a solid experience to replicate data from Postgres to data warehouses. Now we're expanding to queues. PeerDB Streams provides a simple and native way to replicate changes as they happen in Postgres to Queues (Kafka, Redpanda, Google PubSub, etc). We use Postgres logical decoding to enable Change Data Capture (CDC). Blog post here: https://blog.peerdb.io/peerdb-streams-simple-native-postgres... . 10-min quickstart here: https://docs.peerdb.io/quickstart/streams-quickstart . We chose queues as many users found that existing tools are complex. Debezium is the most used tool for this use-case. It has large production usage. However, a common pain point among our users is that it has a significa
Why it failed: Hello HN, I am Sai Srirampur, one of the Co-founders of PeerDB. ( https://github.com/PeerDB-io/peerdb ). We spent the past 7 months building a solid experience to replicate data from Postgres to data warehouses. Now we're expanding to queues. PeerDB Strea
saisrirampur
- startupShut down2024
Show HN: TideCloak – Decentralized IAM for security and user sovereignty
Hey HN! After 6 years of R&D, our small team is excited to share our project TideCloak - an IAM designed to help developers move fast without worrying about catastrophic breaches or overpowered admins with keys to the kingdom. Traditional IAMs rely on centralized authority - admins, root certificates, and decryption keys - which create glaring vulnerabilities in a breach. To address this, we’ve integrated Keycloak (Red Hat’s IAM) with a decentralized key architecture powered by our (academically validated) Ineffable Cryptography. Here’s the idea: keys are split across a decentralized network (our Cybersecurity Fabric) so no one ever holds the full key. Even in a breach or F$%k up, there’s no unchecked authority exposed. Right now, TideCloak uses the Cybersecurity Fabric as an IdP, meaning users authenticate without their credentials being stored or shared. Essentially, users bring th
Why it failed: Hey HN! After 6 years of R&D, our small team is excited to share our project TideCloak - an IAM designed to help developers move fast without worrying about catastrophic breaches or overpowered admins with keys to the kingdom. Traditional IAMs rely on centralized authority -
SaltNHash
- startupShut down2024
Launch HN: Manaflow (YC S24) – Automate repetitive office work in tables
Hey HN! We’re Austin, Lawrence, and Wesley from Manaflow ( https://manaflow.com ). Manaflow is a spreadsheet where each column represents a step instruction of a task and each row represents an AI agent executing a case of the task. With one click, you can execute thousands of tasks involving data retrieval, gluing APIs, and taking actions. Here’s a video demo: https://www.youtube.com/watch?v=jwaaqjHGkT4 And a live demo: https://manaflow.com/demo The idea to build an AI automation tool for small-to-mid-sized businesses (SMBs) emerged from our extensive conversations with local managers, directors, and operators. We learned that SMBs are constantly overwhelmed with manual workflows and heavily underutilize technology, especially compared to bigger businesses with teams of engineers. Witnessing an operations manager show us folders of spreadsheets fo
Why it failed: Hey HN! We’re Austin, Lawrence, and Wesley from Manaflow ( https://manaflow.com ). Manaflow is a spreadsheet where each column represents a step instruction of a task and each row represents an AI agent executing a case of the task. With one click, you can execute thous
austinwang115
- startupShut down2024
Ask HN: Projects You Thought Were Cool but Failed Miserably – What Happened?
Hey HN community, I'm here to spark a discussion on a topic that, I believe, many of us have encountered but might not often talk about: projects that we were excited about and believed in, but that ended up failing miserably after release. To kick things off, I want to share my own experience with a project called interfAIce. It's a Java/Kotlin library that allows developers to access Large Language Models (LLMs) by defining interfaces. The library automatically generates proxies that query the OpenAI API and formats the results into the defined return type data. I thought it was super cool at the time of development. Despite my enthusiasm, my library never gained the attention I hoped for after its release. Reflecting on it, I believe the primary reason might be the mismatch between the technology stacks used by Java/Kotlin developers, who are mostly in server/
Why it failed: Hey HN community, I'm here to spark a discussion on a topic that, I believe, many of us have encountered but might not often talk about: projects that we were excited about and believed in, but that ended up failing miserably after release. To kick things off, I want to shar
mscheong
- startupShut down2024
Ask HN: How do I find a job, in this market with ADHD
Hello HN, A little background about me. I am a software developer, based in US, currently. Due to a lot of personal issues in my life, and upon reflecting about myself about things I faced since my childhood. I visited a psychiatrist recently and was diagnosed with ADHD. I consider myself a good coder, I'm very creative with my approach to the problems. I pick new things very easily and my skillset is also very diverse. My employers in past have always been happy to have me on team. And always consider me an asset. But one thing I really suck at is interviews. My mouth just can't keep up with my mind and I end up blurting nonsense. I mess up the easiest questions. My skillset doesn't translate to the interview format at all. I fiddle, I get lost. Worst part is, I started my career as a freelancer, then I got an offer from a start-up after being impressed with my work. Then
Why it failed: Hello HN, A little background about me. I am a software developer, based in US, currently. Due to a lot of personal issues in my life, and upon reflecting about myself about things I faced since my childhood. I visited a psychiatrist recently and was diagnosed with ADHD. I consid
sillycorn
- startupShut down2024
Launch HN: Parity (YC S24) – AI for on-call engineers working with Kubernetes
Hey HN — we’re Jeffrey, Coleman, and Wilson, and we’re building Parity ( https://tryparity.com ), an AI SRE copilot for on-call engineers working with Kubernetes. Before you've opened your laptop, Parity has conducted an investigation to triage, determine root cause, and suggest a remediation for an issue. You can check out a quick demo of Parity here: https://tryparity.com/demo We met working together as engineers at Crusoe, a cloud provider, and we always dreaded being on-call. It meant a week of putting our lives and projects on hold to be prepared to firefight an issue at any hour of the day. We experienced sleepless nights after being woken up by a PagerDuty alert to then find and follow a runbook. We canceled plans to make time to sift through dashboards and logs in search of the root cause of downtime in our k8s cluster. After speaking with other devs
Why it failed: Hey HN — we’re Jeffrey, Coleman, and Wilson, and we’re building Parity ( https://tryparity.com ), an AI SRE copilot for on-call engineers working with Kubernetes. Before you've opened your laptop, Parity has conducted an investigation to triage, determine root caus
wilson090
- startupShut down2024
Launch HN: Azalea Robotics (YC S24) – Baggage-handling robots for airports
Hey HN! We’re David and John B, cofounders of Azalea Robotics ( https://www.azalearobotics.com ). We build robots to handle passenger baggage in airports. Here are some videos to give you the idea: Unedited autonomous ops: https://www.youtube.com/watch?v=DuJ3ZORnO1o Teleoperated (sped up, so no sound): https://www.youtube.com/watch?v=LeK8NQLnYgA The marketing version: https://www.youtube.com/watch?v=k0SDPm09U6s Robotics is in an interesting place right now, with many warehouse automation companies humming along for almost a decade, and a lot of new effort going to full general purpose hardware with humanoids and software via generalist robotics foundation models. We love these efforts (David used to work on one at Google X with Everyday Robots), but we also see a lot of utility in the current wave of robotics planning and perception
Why it failed: Hey HN! We’re David and John B, cofounders of Azalea Robotics ( https://www.azalearobotics.com ). We build robots to handle passenger baggage in airports. Here are some videos to give you the idea: Unedited autonomous ops: https://www.youtube.com/watch?v=
dmillard
- startupShut down2024
Show HN: Itineraries.io – I built a joint trip planner in between surgeries
Hey HN, For the last several weeks, I have just been sleeping, eating, operating and developing itineraries.io. Rinse and repeat. I work as a surgeon in the UK. My main other passions are travel and programming. Recently, when I haven't been stitching someone up, I've been working on my project. Having always dreamed of exploring the world as a child, I struck a goldmine when I found a wife who shared the same love of adventure as me. We couldn't afford to travel much when we first met at university. Over the last couple of years, since both entering the workforce, we have been able to live out some of our dreams. It has been wonderful. We now have a little one coming along and I can't wait to adventure as a family. I created itineraries.io because my wife and I usually rely on making Excel spreadsheets for our travels. These eventually become quite detailed. I though
Why it failed: Hey HN, For the last several weeks, I have just been sleeping, eating, operating and developing itineraries.io. Rinse and repeat. I work as a surgeon in the UK. My main other passions are travel and programming. Recently, when I haven't been stitching someone up, I've b
samaralihussain
- startupShut down2024
Show HN: ServerlessAI – Build, scale, and monetize AI apps without back end
Hello HN, I’ve always loved building frontend-only apps—those you can prototype over a weekend, host for free on GitHub Pages, and scale to millions of users. Unfortunately, AI-enabled apps complicate things, as exposing your OpenAI key to the world is obviously a no-go. This also means mobile developers often have to run their own servers. That’s why I built ServerlessAI, an API gateway that lets you securely call multiple AI providers directly from client side using OpenAI-compatible APIs. You can authenticate users through any identity provider, like Google or Apple, and set per-user request or spending quotas. You can also define an allowlist for endpoints and models. To monetize, you can apply different quotas for various user tiers. To start, I recommend checking out our tutorials, where we walk you through building a complete, deployment-ready AI app in 5 minutes. We’ve got tutori
Why it failed: Hello HN, I’ve always loved building frontend-only apps—those you can prototype over a weekend, host for free on GitHub Pages, and scale to millions of users. Unfortunately, AI-enabled apps complicate things, as exposing your OpenAI key to the world is obviously a no-go. This als
HeavenFox
- startupShut down2024
Launch HN: Codebuff (YC F24) – CLI tool that writes code for you
Hey HN! We’re James and Brandon building Codebuff ( https://codebuff.com ). Codebuff is like Cursor Composer, but in your terminal: it modifies files based on your natural language requests. You can try it with `npm i -g codebuff` and start using it immediately for free. We have no login gate, and we give all accounts up to $20 worth of credits. Codebuff is different because we simplified the input to one step: you type what you want done in your terminal and hit enter. Then Codebuff looks through your whole codebase and makes the edits it wants, to existing source files or new ones. It also can run your tests, the type checker, or install packages to fulfill your request. Demo video: https://www.youtube.com/watch?v=dQ0NOMsu0dA It all started at a hackathon. I was trying out Sonnet 3.5 which had recently come out and seeing if I could use it to write code. The sc
Why it failed: Hey HN! We’re James and Brandon building Codebuff ( https://codebuff.com ). Codebuff is like Cursor Composer, but in your terminal: it modifies files based on your natural language requests. You can try it with `npm i -g codebuff` and start using it immediately for free
jahooma
- startupShut down2024
Show HN: I wrote a partial re-implementation of DirectMusic
DirectMusic[1] is a deprecated Windows API for playing dynamic music scores authored using DirectMusic Producer[2]. It was originally released as part of DirectX in 1999 and discontinued with the release in Windows Vista around 2009-ish. Due to my involvement with projects[3] re-implementing an old game engine for the early 2000's games Gothic and Gothic II[4], I came to notice that existing solutions[5] were incorrect and hard to use. Thus, I was tasked with writing a new, correct re-implementation of the API. Today, my re-implementation is able to (mostly) play back so-called style-based segments[6] and is fully tested against both the Gothic and Gothic II soundtracks. I am actively working on getting the Lego Island 2 soundtrack working as well. There are many features of DirectMusic which the library does not currently support, simply because I have not been able to find or test
Why it failed: DirectMusic[1] is a deprecated Windows API for playing dynamic music scores authored using DirectMusic Producer[2]. It was originally released as part of DirectX in 1999 and discontinued with the release in Windows Vista around 2009-ish. Due to my involvement with projects[3] re-
lmichaelis
- startupShut down2024
Show HN: Pg_mooncake – Delta/Iceberg columnstore tables in Postgres
Hello HN, one of the founders of Mooncake Labs here. Today, we are launching pg_mooncake, an extension that brings columnstore tables with DuckDB execution to Postgres. Expect analytical performance akin to DuckDB on Parquet (clickbench results to come soon). You can run transactional updates, deletes, inserts on these tables, and they're written as Delta Lake tables (and soon Iceberg) to your object store (S3, etc.). pg_mooncake is live on Neon today. Let us know what you think. It'll be coming to Supabase shortly, and other Postgres providers in the future. Why another postgres analytics extension? We actually leverage pg_duckdb and DuckDB as our execution engine — this is how we were able to ship the extension in just 60 days. We wrote about this with the DuckDB team: https://motherduck.com/blog/pg-mooncake-columnstore/ pg_mooncake shines in two scen
Why it failed: Hello HN, one of the founders of Mooncake Labs here. Today, we are launching pg_mooncake, an extension that brings columnstore tables with DuckDB execution to Postgres. Expect analytical performance akin to DuckDB on Parquet (clickbench results to come soon). You can run transact
sunzhousz
- startupShut down2024
Show HN: Control webviews programmatically with the Pipeable iOS SDK
For the past few months we have been working on the PipeableSDK and today we are excited to show it to you all! What is Pipeable? Pipeable is an open-source iOS and (soon) Android SDK that allows developers to programmatically control webviews in mobile apps. Webviews are essentially in-app browsers that display web content directly within the app. If you are familiar with Puppeteer or Playwright, Pipeable serves a similar purpose but for these embedded browsers, offering granular control over web interactions. What can you use Pipeable for? You can use Pipeable to automate web interactions in your mobile app. This can be anything from scraping data from a website, to automating a user’s interaction with a website, to creating a mobile app that acts as an agent on behalf of a user. You can check out https://github.com/PipeableHQ/pipeable-ios-samples for some concrete
Why it failed: For the past few months we have been working on the PipeableSDK and today we are excited to show it to you all! What is Pipeable? Pipeable is an open-source iOS and (soon) Android SDK that allows developers to programmatically control webviews in mobile apps. Webviews are essenti
deyan
- startupShut down2024
Ask HN: Stripe Atlas messed up my 83B election what do I do? [fixed]
Formed my new startup with Stripe Atlas which had worked well in the past. I used their auto 83b feature to mail in our 83b elections, the UI updated and said everything was good. I received an email update > Great news! Your 83(b) election was filed on time. I go to get a copy of my EIN letter and notice that the atlas UI has updated to: > Your IRS 83(b) election deadline has passed. > We’ve noticed it’s taking longer than normal for USPS to deliver your 83(b) election. Not to worry, you’ve successfully filed on time (it’s based on the postmark date, not delivery date). The Proof of Timely Election letter and USPS tracking site both say the letter is "In Transit To Next Facility, Arriving Late". This letter was sent over a month ago. What are my options now? * Contact the IRS to see if it was delivered * Continue with my equity in a non-tax advantaged status * Dissolv
Why it failed: Formed my new startup with Stripe Atlas which had worked well in the past. I used their auto 83b feature to mail in our 83b elections, the UI updated and said everything was good. I received an email update > Great news! Your 83(b) election was filed on time. I go to get a cop
_h9vu
- startupShut down2024
Lfgss shutting down 16th March 2025 (day before Online Safety Act is enforced)
figured this might be interesting... I run just over 300 forums, for a monthly audience of 275k active users. most of this is on Linode instances and Hetzner instances, a couple of the larger fora go via Cloudflare, but the rest just hits the server. and it's all being shut down. the UK Online Safety Act creates a massive liability, and whilst at first glance the risk seems low the reality is that moderating people usually provokes ire from those people, if we had to moderate them because they were a threat to the community then they are usually the kind of people who get angry. in 28 years of running forums, as a result of moderation I've had people try to get the domain revoked, fake copyright notices, death threats, stalkers (IRL and online)... as a forum moderator you are known, and you are a target, and the Online Safety Act creates a weapon that can be used against you. t
Why it failed: figured this might be interesting... I run just over 300 forums, for a monthly audience of 275k active users. most of this is on Linode instances and Hetzner instances, a couple of the larger fora go via Cloudflare, but the rest just hits the server. and it's all being shut
buro9
- startupShut down2024
Show HN: I'm a doctor and built a responsive breathing app for anxiety and sleep
Hey HN! I’m an NHS doctor and the founder of Pia ( https://www.piahealth.co ) which developed Lungy ( https://www.lungy.app ). Lungy is an iOS app that responds to breathing in real-time and was designed to make breathing exercises more engaging and beneficial to do. It’s been just over a year since Lungy launched (here’s the original ShowHN: https://news.ycombinator.com/item?id=34534615 ) and it's had a huge update and complete redesign. We rebuilt the whole app, and added a real-time 3D soft body solver which gives some really cool interactions like blobs / objects that inflate as you breathe. We also made a version for Vision Pro, called 'Lungy Spaces'. My background is as a junior surgical trainee and I started building Lungy in 2020 during the first COVID lockdown in London. During COVID, there were huge numbers of patients comi
Why it failed: Hey HN! I’m an NHS doctor and the founder of Pia ( https://www.piahealth.co ) which developed Lungy ( https://www.lungy.app ). Lungy is an iOS app that responds to breathing in real-time and was designed to make breathing exercises more engaging and beneficial
lukko
- startupShut down2024
Ask HN: What's a security risk you think people should be more aware of?
Here's mine: QR codes. They're a terrible idea, especially when there's no URL attached. They remove someone's ability to verify that yes, this is the correct site, and are easy to bypass - a scammer can literally just put a different QR code over the intended one, and ta-da! Done.
Why it failed: Here's mine: QR codes. They're a terrible idea, especially when there's no URL attached. They remove someone's ability to verify that yes, this is the correct site, and are easy to bypass - a scammer can literally just put a different QR code over the intended
Yawrehto
- startupShut down2024
Show HN: PlayBooks – Jupyter Notebooks style on-call investigation documents
Hello everyone, Dipesh and Siddarth here. We are building PlayBooks ( https://github.com/DrDroidLab/playbooks ), an open source tool to write executable notebooks for on-call investigations / remediations instead of Google Docs or Wikis. There’s a demo video here: https://www.youtube.com/watch?v=_e-wOtIm1gk , and our docs are here: https://docs.drdroid.io/docs/playbooks We were in YC’s W23 batch working on a data lakehouse with support for dynamic log schemas. Eventually we realized it was a product in search of a market and decided to stop building it. When pivoting, we decided to work on something that we originally prototyped (before even YC) but didn’t execute on. In our previous jobs, we were at a food delivery startup in India with a busy on-call routine for backend & devops engineers and a small tech team. Often busin
Why it failed: Hello everyone, Dipesh and Siddarth here. We are building PlayBooks ( https://github.com/DrDroidLab/playbooks ), an open source tool to write executable notebooks for on-call investigations / remediations instead of Google Docs or Wikis. There’s a demo vi
TheBengaluruGuy
- startupShut down2024
Postgres.new: In-browser Postgres with an AI interface
hey HN, supabase ceo here This is a new service that we're experimenting with that uses PGLite[0], a WASM build of Postgres that runs in the browser. You might remember an earlier WASM build[1] that was around ~30MB. The Electric team [2] have gone one step further and created a complete build of Postgres that’s under 3MB. Their implementation is technically interesting. Postgres is normally multi-process - each client connection is handed to a child process by the postmaster process. In WASM there’s limited/no support for process forking and threads. Fortunately, Postgres has a relatively unknown built-in “single user mode” [3] primarily designed for bootstrapping a new database and disaster recovery. Single-user mode only supports a minimal cancel REPL, so PGlite adds wire-protocol support which enables parametrised queries etc. We have created https://postgres.new
Why it failed: hey HN, supabase ceo here This is a new service that we're experimenting with that uses PGLite[0], a WASM build of Postgres that runs in the browser. You might remember an earlier WASM build[1] that was around ~30MB. The Electric team [2] have gone one step further and creat
kiwicopple
- startupShut down2024
Launch HN: GPT Driver (YC S21) – End-to-end app testing in natural language
Hey HN, we are Chris and Chris from MobileBoost ( https://mobileboost.io/ ). We’re building GPT Driver, an AI-native approach to create and execute end-to-end (E2E) tests on mobile applications. Our solution allows teams to define tests in natural language and prevents test flakiness by taking a visual approach paired with LLM (Large Language Model) reasoning. This helps achieve E2E test coverage with a fraction of the usual effort. You can watch a brief product walkthrough here: https://www.youtube.com/watch?v=5-Ge2fqdlxc In terms of trying the product out: since the service is resource-intensive (we provide hosted virtual/real phone instances), we don't currently have a playground available. However, you can see some examples here https://mobileboost.io/showcases and book a demo of GPT Driver testing your app through our website. W
Why it failed: Hey HN, we are Chris and Chris from MobileBoost ( https://mobileboost.io/ ). We’re building GPT Driver, an AI-native approach to create and execute end-to-end (E2E) tests on mobile applications. Our solution allows teams to define tests in natural language and prev
cschiller
- startupShut down2024
We Built a Tool to Hack Our Own AI: Lessons Learned Securing Chatbots/Voicebots
Hey HN, We’ve been working in-house on a platform that tests the security of chatbots and voicebots by intentionally trying to break them. As AI-driven bots become more prevalent across sectors like customer service, healthcare, and finance, ensuring they are secure from exploitation is critical. Many companies focus on training their AI to perform well but often overlook the necessity of breaking them to identify vulnerabilities—essential to ensuring their robustness in the real world. Why We Built This: We realized how easily AI models could be manipulated through adversarial inputs and social engineering tactics. With the rise of chatbots and voicebots in sensitive areas, traditional testing methods fell short. What We Did: We developed an in-house platform (code named RedOps) that simulates real-world attacks on chatbots and voicebots, including: 1. Contextual Manipulation: Testing h
Why it failed: Hey HN, We’ve been working in-house on a platform that tests the security of chatbots and voicebots by intentionally trying to break them. As AI-driven bots become more prevalent across sectors like customer service, healthcare, and finance, ensuring they are secure from exploita
titusblair
- startupShut down2024
Show HN: Neosync – Open-Source Data Anonymization for Postgres and MySQL
Hey HN, we're Evis and Nick and we're excited to be launching Neosync ( https://www.github.com/nucleuscloud/neosync ). Neosync is an open source platform that helps developers anonymize production data, generate synthetic data and sync it across their environments for better testing, debugging and developer experience. Most developers and teams have some version of a database seed script that creates some mock data for their local and stage databases. The problem is that production data is messy and it’s very difficult to replicate that with mock data. This causes two big problems for developers. The first problem is that features seem to work locally/stage but have bugs and edge cases in production because the seed data you used to develop against was not representative of production data. The second problem we saw was that debugging production errors
Why it failed: Hey HN, we're Evis and Nick and we're excited to be launching Neosync ( https://www.github.com/nucleuscloud/neosync ). Neosync is an open source platform that helps developers anonymize production data, generate synthetic data and sync it across thei
edrenova
- startupShut down2024
Show HN: Route your prompts to the best LLM
Hey HN, we've just finished building a dynamic router for LLMs, which takes each prompt and sends it to the most appropriate model and provider. We'd love to know what you think! Here is a quick(ish) screen-recroding explaining how it works: https://youtu.be/ZpY6SIkBosE Best results when training a custom router on your own prompt data: https://youtu.be/9JYqNbIEac0 The router balances user preferences for quality, speed and cost. The end result is higher quality and faster LLM responses at lower cost. The quality for each candidate LLM is predicted ahead of time using a neural scoring function, which is a BERT-like architecture conditioned on the prompt and a latent representation of the LLM being scored. The different LLMs are queried across the batch dimension, with the neural scoring architecture taking a single latent representation of the LLM
Why it failed: Hey HN, we've just finished building a dynamic router for LLMs, which takes each prompt and sends it to the most appropriate model and provider. We'd love to know what you think! Here is a quick(ish) screen-recroding explaining how it works: https://youtu.be&#
danlenton