ege = {
"role": "AI Engineer & Full-Stack Dev",
"focus": ["LLM Automation", "Self-hosted AI", "Open Source"],
"stack": ["Python", "TypeScript", "Docker", "Linux"],
"affiliation": "INOVENS · Hackathons · Client Projects",
"building": ["TrustLaya-S", "mcp-forge", "mcp-sanity"],
"open_source": "84 PRs across 36 upstream repos, October 2026",
"motto": "Automate everything. Ship fast. Keep it simple."
}84 PRs · 36 upstream repos · 21 merged · 33 in review · October 2026
Merged
In review — selected open PRs
| PR | Repo | Change |
|---|---|---|
| #18656 | ollama/ollama (181k⭐) |
Use UUIDs for OpenAI-compatible response ids |
| #18608 | ollama/ollama (181k⭐) |
Support max_completion_tokens as a generation cap |
| #18520 | ollama/ollama (181k⭐) |
Allow 96-char model names so long HF GGUF pulls validate |
| #58341 | vllm-project/vllm (92k⭐) |
Guard /invocations with the API key check |
| #57518 | vllm-project/vllm (92k⭐) |
LoRA device lookup falls through to packed weights when repacking clears… |
| #8248 | microsoft/autogen (61k⭐) |
Carry JSON Schema type for optional tool fields |
| #43145 | BerriAI/litellm (59k⭐) |
fix(utils): pass allowed_tools tool_choice through to providers |
| #41572 | BerriAI/litellm (59k⭐) |
fix(proxy): guard SlackAlerting periodic_flush start on repeated update_… |
| #41243 | BerriAI/litellm (59k⭐) |
fix(passthrough): keep adaptive thinking/effort on native /v1/messages p… |
| #41012 | BerriAI/litellm (59k⭐) |
fix(proxy): case-insensitive User-Agent header lookup in _get_user_agent… |
| #2834 | debpalash/VoiceStudio (50k⭐) |
fix(subtitles): strip ruby pronunciation annotations from spoken text |
| #1699 | alibaba/open-code-review (46k⭐) |
fix(session): return empty slice for nonexistent sessions directory |
Plus open work in codex-switcher, Claude-Red, whisper-asr-webservice, astronomer-cosmos, codex-switcher, continue, dagster, 9router, ddgs, deepseek-recipe, langfuse, mcp-agent, whatsapp-mcp, typescript-sdk, ray, memtrace-public.




