Skip to content
View serendip-ml's full-sized avatar
💭
Building!
💭
Building!
5 followers · 15 following

Achievements

Achievement: Pull Sharkx4Achievement: YOLOAchievement: QuickdrawAchievement: Pair Extraordinaire

Achievements

Achievement: Pull Sharkx4Achievement: YOLOAchievement: QuickdrawAchievement: Pair Extraordinaire

Organizations

@llm-works

Block or report serendip-ml

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
serendip-ml/README.md

Production AI agent infrastructure

AI is systems engineering. The hardest part of production AI agents is the infrastructure around the models: off-LLM guarantees, verification loops, response guards, per-role model routing.

🔥 Live products

📰 news.llm-works.ai: AI news aggregation across 70+ sources with theme clustering, daily digests, and persona-based presentation

🔬 xray.llm-works.ai: Decision-grade deep research with calibrated verdicts through adversarial + hypothesis + grader verification loops, plan-as-protagonist architecture, knowledge-graph synthesis

📚 Open source stack

Package What it does
🤖 llm-gent Trait-based agent framework with production HTTP runtime
🧩 llm-saia Verb-vocabulary LLM interaction (ask · verify · critique · refine · ...)
🧠 llm-kelt Persistent memory + embeddings + LoRA/DPO adapter training (Postgres+pgvector)
🔌 llm-infer Unified inference (vLLM · Ollama · native torch), multi-provider client
🔩 appinfra Production Python infrastructure (logging · DB · lifecycle)

📊 ~500K lines across all packages · 🧪 ~1:1 test-to-code ratio · 💻 Local Qwen3.5-27B on RTX PRO 6000

📝 Blog · 🌐 llm-works.ai · 🏢 github.com/llm-works

Pinned Loading

  1. llm-works/llm-gent llm-works/llm-gent Public

    Composable agent framework with trait-based architecture and production HTTP runtime

    Python

  2. llm-works/llm-saia llm-works/llm-saia Public

    Structured verbs for LLM interactions (verify, critique, decompose, extract, synthesize...)

    Python 1

  3. llm-works/llm-kelt llm-works/llm-kelt Public

    Framework for collecting and managing LLM context: facts storage, feedback collection, RAG retrieval, and LoRA fine-tuning

    Python 1

  4. llm-works/llm-infer llm-works/llm-infer Public

    LLM inference server with native, vLLM, and Ollama backends, including a client to interact with multiple LLM providers

    Python

  5. llm-works/appinfra llm-works/appinfra Public

    Production-grade Python infrastructure framework for building reliable CLI tools and services

    Python

  6. claude-code-skills claude-code-skills Public

    Evidence-based code analysis skills for Claude Code: project grading and pre-PR review