Real-world
Coding
Concrete workflows, including the tradeoffs and limitations reported by their authors.
this collection
Building a full SaaS product on Qwen 3.8 27B with vLLM
Switching from llama.cpp to vLLM on AMD hardware was the unlock for running a complete SaaS build loop through a local model.
Shipping seven Kotlin Multiplatform projects on a fully custom local harness
Not vibe coding: a custom harness on a patched llama.cpp build has shipped seven real KMP projects.
Coding a SaaS product on an M1 Max, one well-scoped feature at a time
Running Qwen 3.8 27B on the same Mac they work on, this developer credits per-feature domain glossaries and tight scoping — not the model — for making local coding reliable.
Full app development on a single 3090 with OpenCode
A four-bit Qwen 3.8 quant at 200K context handles complete app builds through OpenCode, left running until it hits a decision point.
An RTX 6000 Pro and an M4 Max, benchmarked against an active Claude Max plan
This person kept paying for Claude Max specifically so they'd have a real comparison point for their own coding hardware — and admits the math still doesn't favor local.
A context-aware router working on the first try
A router in front of compiled llama.cpp servers, feeding Qwen 3.8 27B through OpenCode with downloaded skills, worked essentially out of the box.
98% local, on a self-extended Pi Agent 'software factory'
DeepSeek V4 Flash Vision on two DGX Sparks, wrapped in a heavily extended Pi Agent setup, handles nearly all of this person's hands-off work.
Eight A100s in tensor-parallel, inside a company with a hundred more
A large enterprise deployment: eight A100 80GBs running Qwen 3.8 under SGLang with tensor parallelism, while the company brings newer B200s online for bigger models.
The first local model this person actually trusts with real work
On a 12GB 5070, a heavily quantized Qwen 3.8 27B through a custom Hermes backend is, in their words, the first local model that's felt confidently productive.
Switching back a model version to escape endless self-checking loops
Qwen 3.8 kept spiraling into overlong, indecisive reasoning on this person's coding tasks, no matter the configuration — so they moved back to Qwen 3.6.
Using a local model to stress-test the agent a coding agent is building
An unusual layered setup: the local model isn't the coding agent — it's the thing an outer coding agent uses to build and stress-test a separate 'inner' agent.
Background coding at 10-15 tok/s on an Intel Arc laptop GPU
A NUC12 with a 16GB Intel Arc A770M handles a 262K-context Qwen build entirely in VRAM, coding in the background while the owner does something else.
Rotating between three harnesses to find what actually works
After trying Claude Code with task-farming and a Qwen-as-orchestrator experiment, this person settled into two parallel workflows — Orca as the local workhorse, and OpenCode or Bionic for solo agent development.
Good enough for 80% of coding, honest about the other 20%
Daily local coding on a 4070 is explicitly not as good as Claude for complex refactoring — but the privacy and iteration speed still make it worth keeping.