CJ (Coding Garden)

Senior Creator at @syntaxfm / @sentry

Host of Coding Garden on Twitch and YouTube

Co-organizer of @DenverScript

[Syntax @syntaxfm]

We just sent $100,000 to 20 open source projects ($5000 each) that the Syntax team use regularly and depend on. Thank you so much to all the maintainers for all your time and efforts!

Big thanks to @sentry and @ThePledge for giving us the resources to do this.

https://t.co/nf7LgkKEqD

[Syntax @syntaxfm]

We are live to chat about how Deno is dead, SubQ Snap is better than Jev, OpenCode 2 released, Tanstack Start 1.0 is stable and more!

See you in the chat.

https://t.co/qiW7wrfaXU

https://t.co/SW6LGXtODa

[CJ (Coding Garden) @CodingGarden]

@developerzeke @stolinski @wesbos 🙅‍♂️

[CJ (Coding Garden) @CodingGarden]

@ozzyonfire It's a custom sega game.

https://t.co/ilSwo7aoZI

[CJ (Coding Garden) @CodingGarden]

@spacebars_tabs @syntaxfm Yes we do. https://t.co/DIm1qpBVOc

https://t.co/Nn3lbwKbTU

https://t.co/mV79VD6q2j

https://t.co/mzmpvMGtHS

[CJ (Coding Garden) @CodingGarden]

What open source projects should we include?

[Syntax @syntaxfm]

We have $100K to give away to open source maintainers! Let us know your suggestions and nominations.

[CJ (Coding Garden) @CodingGarden]

@waybackmachine the wayback machine page is firing thousands of requests (eventually getting 429) when I scroll the calendar into view for a URL.

[CJ (Coding Garden) @CodingGarden]

I made a Syntax themed Game Boy Color game.

https://t.co/FgRqQQb5zY

[Syntax @syntaxfm]

The Syntax crew is LIVE to chat how Ruby on Rails is Dead, Zuck cooked with Muse, upm the tiny npm replacement and more!

See you in the chat.

https://t.co/NUeKnKwxmQ

https://t.co/37S39iLwkM

https://t.co/JgG1HmNixD

[CJ (Coding Garden) @CodingGarden]

@stolinski Check out @omidotme fully open source and they have a dev kit https://t.co/n6JaWDFy5O

[Syntax @syntaxfm]

The Syntax crew is LIVE to chat about Jev! Is the Hype Overblown? Also Claude shipped AGENTS.md support, TinyCast is a smaller Raycast and more!

See you in the chat.

https://t.co/qiW7wrfaXU

https://t.co/aeFNck77Iv

[CJ (Coding Garden) @CodingGarden]

Code for jev chat is available here: https://t.co/T2bstTvfhA

The README includes a breakdown of how this all works and includes an FAQ.

[CJ (Coding Garden) @CodingGarden]

I built a chat bot with jev, no LLM at all! Responses are instant, no hallucinations.

I hooked it up to web search, wikipedia, weather, todoist and home assistant.

Jev decides what tool to call and what args to use based on the prompt. Instant answers cite sources as well!

[CJ (Coding Garden) @CodingGarden]

@offchan420 @49agents Requires an extra request to resolve spelling errors, but still works. Combined with cspell and compromise nlp pre-processing, jev can decide what the user actually meant.

[CJ (Coding Garden) @CodingGarden]

I built a chat bot with jev, no LLM at all! Responses are instant, no hallucinations.

I hooked it up to web search, wikipedia, weather, todoist and home assistant.

Jev decides what tool to call and what args to use based on the prompt. Instant answers cite sources as well!

[CJ (Coding Garden) @CodingGarden]

@ethereaglehq The answer is directly from the article right now.

It detects a wikipedia lookup. Gets the article. Then does a second request that includes the full article where each sentence is an option for jev to choose.

See their guide here: https://t.co/lobNkRX59A

[CJ (Coding Garden) @CodingGarden]

I'd have to create a new "timer" tool and make it available to jev but definitely doable! Jev would just respond with which action to time delay, and how long. Then we pass that off to a background queue that invokes it when necessary.

Compound requests are a bit tricky though. The smart home example from typesafe AI actually detects a compound request, passes to an llm to de-compose into individual commands. Then passes each command to jev.

[CJ (Coding Garden) @CodingGarden]

@LifeOfPhat Nope! That's just a wikipedia lookup. Jev extracts the topic the user is asking about, queries wikipedia, then a follow up is sent to jev with the content of the wikipedia article asking which line from the article answers the user's prompt.

[CJ (Coding Garden) @CodingGarden]

I pass a "choice" to jev - "Which tool would fulfil the user's latest request (taking the conversation into account)?" and the options are a dynamic list of every available tool.

For arg extraction / matching on the user prompt there is a seperate question for every single available tool like: "For a weather request: which city or place does the user want the weather for? If they don't name one, pick the place from the earlier request." and the options provided are extracted from the user prompt as a list of options. Could be the prompt split on spaces, could be smarter word pairs.

[CJ (Coding Garden) @CodingGarden]

I provide an option to jev for home assistant "no device or room is mentioned at all" - this then asks for a follow up that extracts and matches the name to a device in the home.

We can do a similar thing for ambiguous tool calls. Provide jev an option "More than one tool call can satisfy this request" and then give a follow up prompt to choose, OR detect high match percent in multiple tools and automatically give a follow up to the user before executing.

[CJ (Coding Garden) @CodingGarden]

Watch the full explainer: https://t.co/6wD393fGPl