DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Turning Qwen2.5-0.5B into a JSON API with SFT — 23% 100% on a Free T4

Turning Qwen2.5-0.5B into a JSON API with SFT — 23% 100% on a Free T4

1
Comments
13 min read
I benchmarked 7 Python JSON parsers on 300 malformed LLM outputs — mine lost the column I cared about

I benchmarked 7 Python JSON parsers on 300 malformed LLM outputs — mine lost the column I cared about

Comments
7 min read
Building Kisan Mitra: How I Built an Ultra-Fast Voice AI for Indian Farmers in 10 Days

Building Kisan Mitra: How I Built an Ultra-Fast Voice AI for Indian Farmers in 10 Days

Comments
4 min read
Moderation Report Triage: Node.js LLM API JSON Contracts for Portable Summaries

Moderation Report Triage: Node.js LLM API JSON Contracts for Portable Summaries

Comments
5 min read
My support agent answered 9 of 12 tickets. The other 3 are the point.

My support agent answered 9 of 12 tickets. The other 3 are the point.

Comments
3 min read
Learn to Budget a Free Model Tier by Building a Tiny Token Ledger

Learn to Budget a Free Model Tier by Building a Tiny Token Ledger

Comments
3 min read
The 7.4% You Don't See: Checkpointing Long LLM Jobs Before They Time Out

The 7.4% You Don't See: Checkpointing Long LLM Jobs Before They Time Out

1
Comments
4 min read
I Built a World Where the Canon Is Written by AI Agents — 13 Artifacts, 5 LLMs, 0 Human Gatekeepers

I Built a World Where the Canon Is Written by AI Agents — 13 Artifacts, 5 LLMs, 0 Human Gatekeepers

Comments
2 min read
Your RAG pipeline is bad at docs because your chunker splits code fences

Your RAG pipeline is bad at docs because your chunker splits code fences

2
Comments 1
8 min read
Why GPT-5.6 Luna High Is My Default for Agentic Engineering

Why GPT-5.6 Luna High Is My Default for Agentic Engineering

Comments
4 min read
Silent Regressions Have No Stack Trace: A Minimal Prompt Eval Harness

Silent Regressions Have No Stack Trace: A Minimal Prompt Eval Harness

Comments 2
5 min read
Cline in production: the autonomous code agent for VS Code I use with deliberate constraints

Cline in production: the autonomous code agent for VS Code I use with deliberate constraints

1
Comments
8 min read
Not Every LLM Needs vLLM: A Kubernetes Engineer's Guide to Serving Engines

Not Every LLM Needs vLLM: A Kubernetes Engineer's Guide to Serving Engines

1
Comments 1
12 min read
LLM observability: tracing, monitoring, and debugging agents in production

LLM observability: tracing, monitoring, and debugging agents in production

Comments
11 min read
Run a Pre-Mortem on That Free 30M-Token Allowance Before You Treat It as Headroom

Run a Pre-Mortem on That Free 30M-Token Allowance Before You Treat It as Headroom

Comments
5 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.