Spaces:

lebiraja
/

customer-support-env

Sleeping

App Files Files Community

customer-support-env

3.94 MB

Ctrl+K

Ctrl+K

8 contributors

History: 113 commits

lebiraja's picture

fix(docker): install musl so prebuilt llama-cpp-python wheel loads on Debian

c4d88c8 20 days ago

.obsidian
fix: Phase 17 final production lock (auth, PII redact, structlog) about 1 month ago
colab
fix(colab): 3 bugs — prompt KeyError, response→respond normalize, max_length warning 20 days ago
docs
docs: reorganize .md files into docs/ and rewrite README for Round 2 22 days ago
env
feat: local judge server (serve_judge.py) — replaces NVIDIA NIM entirely 21 days ago
frontend
feat: wire trained GGUF model (customer-support-grpo-v5) into inference 20 days ago
server
feat: wire trained GGUF model (customer-support-grpo-v5) into inference 20 days ago
tests
chore(round2): full audit pass — harden rewards, patch frontend, tighten training 21 days ago
train
fix(train): v5 — JSON preprocessor + dual-token for friend's account 20 days ago
unsloth_compiled_cache
feat:model infrence to local qwen for grpo training 22 days ago
.dockerignore

23 Bytes
feat: complete hackathon hardening and nightmare mode integration about 1 month ago
.gitattributes

42 Bytes
fix(assets): move plot PNGs to frontend/public, remove binary files from results/ 20 days ago
.gitignore

122 Bytes
fix(plots): serve images via GitHub raw URLs, restore results/ for origin 20 days ago
Dockerfile

2.5 kB
fix(docker): install musl so prebuilt llama-cpp-python wheel loads on Debian 20 days ago
README.md

35.7 kB
feat: wire trained GGUF model (customer-support-grpo-v5) into inference 20 days ago
blog.md

19.2 kB
fix: add blog.md at repo root for hackathon submission URL 20 days ago
docker-compose.yml

2.47 kB
feat: local judge server (serve_judge.py) — replaces NVIDIA NIM entirely 21 days ago
inference.py

21.4 kB
fix: natural conversation flow — greet before querying DB 21 days ago
nginx.conf

936 Bytes
fix(space): serve Next.js frontend via nginx reverse proxy 20 days ago
openenv.yaml

11.1 kB
chore(round2): full audit pass — harden rewards, patch frontend, tighten training 21 days ago
pyproject.toml

1.06 kB
feat: hackathon improvements — benchmark dashboard, GRPO training pipeline, grader fixes 23 days ago
requirements.txt

254 Bytes
fix(docker): use prebuilt llama-cpp-python CPU wheel, skip source compile 20 days ago
run_benchmark.py

32.5 kB
feat: pre/post benchmark script + blog.md source 20 days ago
serve_inference.py

4.07 kB
feat: wire trained GGUF model (customer-support-grpo-v5) into inference 20 days ago
serve_judge.py

3.84 kB
feat: local judge server (serve_judge.py) — replaces NVIDIA NIM entirely 21 days ago
start.sh

775 Bytes
feat: wire trained GGUF model (customer-support-grpo-v5) into inference 20 days ago
start_training.sh

11.6 kB
fix(config): update to v5 repo names 20 days ago
stress_test.py

2.91 kB
feat: complete hackathon hardening and nightmare mode integration about 1 month ago
submit_training_job.py

9.47 kB
fix(config): update to v5 repo names 20 days ago
test_complete.sh

1.55 kB
feat: complete hackathon hardening and nightmare mode integration about 1 month ago
test_features.sh

1.91 kB
feat: complete hackathon hardening and nightmare mode integration about 1 month ago
test_live.py

3.17 kB
feat: Round 2 — Hierarchical multi-agent environment (v2.0) 23 days ago
uv.lock

840 kB
feat: hackathon improvements — benchmark dashboard, GRPO training pipeline, grader fixes 23 days ago