frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Ask HN: Why use Jev/decision models at all?

5•niceguy1827•5h ago
Honest question. Jev seems to cost 10x as much as Deepseek and I can easily just put the same payload to Deepseek and add "follow this strict format ...". What am I missing?

Comments

akras14•5h ago
It only makes sense if traditional LLMs are either too expensive or too slow.

That’s it, that’s the only use case.

However there are use cases that meet that requirement, for example, near instant labeling of dynamically generated content on the screen.

runeblaze•5h ago
posttrained models for a specific purpose usually outperforms model that are not posttrained for that purpose
mikericar•4h ago
Check Deterministic State Machine idea - basically, on prod regulated / high stakes env. you don't want LLM to run almost anything directly.

You rather need exact reproducebale, tracable path of decisions. Either a hard-coded graph where LLM is a point of contact for the customer/user and parsing, function call etc. is all done deterministic way.

OR you use smth like JEV to decide about the next path in the chain (under the condition that JEV is trained for that path only and its only job is to do the best routing possible with close to 0 costs).

Tell HN: Major Microsoft Outlook/Exchange Online/Teams Outage in the Americas

8•wmchen•3h ago•1 comments

Tell HN: Cast('abc' AS INTEGER) returns 0 in SQLite

3•prmph•7h ago•3 comments

Ask HN: How do you market your consumer product

5•mv4•5h ago•2 comments

Ask HN: Why use Jev/decision models at all?

5•niceguy1827•5h ago•3 comments

Inception's Diffusion LLM API Is Shutting Down

12•_matthew_•5h ago•6 comments

Ask HN: What are you working on? (November 2026)

299•david927•4d ago•1068 comments

Tell HN: I'm working on original approaches to tag-based file organization

2•prmph•8h ago•1 comments

Claude Code's API billing depends on how often the cache expires

2•MaxBozhenko•8h ago•0 comments

Ask HN: What is holding back robots from being deployed everywhere?

6•mjs06•9h ago•14 comments

ChatGPT Desktop forwards your SSH agent to remote machines

219•DenGorbachev•3d ago•79 comments

Ask HN: Do you still enjoy coding in C++?

61•hasley•3d ago•89 comments

Ask HN: What is your current AI stack looking like?

2•salleisha•5h ago•2 comments

Ask HN: Personal "Tears in the Rain" Moments

14•HoldOnAMinute•1d ago•14 comments

Any good ways on how to get into kernel dev?

7•ihms•1d ago•8 comments

How do you keep up quality when you vibe-code?

37•nkcodes•2d ago•37 comments

What are your favorite tools?

13•d675•1d ago•9 comments

Ask HN: Is lack of human imagination now the bottleneck instead of intelligence?

8•amichail•1d ago•13 comments

Ask HN: How did you convince family and friends to join your XMPP/Matrix server?

102•mtlynch•5d ago•97 comments

Do 9/10 new AI SaaS websites seem like they should be an editor command to you?

5•Kevin_Flynn•1d ago•2 comments

Ask HN: What do your technical interviews look like today? (Nov 2026)

7•Heap4112•21h ago•3 comments

Ask HN: "No AI" apps/companies/resources?

8•BrunoBernardino•1d ago•8 comments

Ask HN: How do you evaluate the skills your AI agents use?

23•enesalatas•3d ago•6 comments

Ask HN: Did anybody used F# in a production?

5•vad_n__•1d ago•3 comments

Ask HN: Software engineer with broken arm, any tips?

6•spapas82•1d ago•11 comments

Tell HN: Another GitHub Major Outage

17•sergiotapia•3d ago•7 comments

Ask HN: Anybody using their own AI harness as well?

5•madprops•1d ago•4 comments

Tell HN: A flood of reports of Opus 5.5 being nerfed

32•b--l•3d ago•17 comments

CoderPad Reports a Security Breach

9•werrett•2d ago•1 comments

uBlock Origin Can't Block Facebook Ads

13•anidsiam•3d ago•5 comments

Why does Reddit want me to use the iOS app on my desktop?

10•corroclaro•1d ago•5 comments