$ whoami
Manas Dani — AI Engineer | MS Data Science @ Indiana University Bloomington
$ cat current_mission.txt
🔭 Building production-grade AI agents with LangGraph, LangChain & RAG
🌱 Diving deep into LLM evaluation, observability & agent architectures
⚡ Making AI measurable, trustworthy, and scalable for the real world
✍️ Writing about Agentic AI on Medium
📫 danimanas28@gmail.com
$ ls skills/
agentic_systems/ rag_pipelines/ llm_evaluation/ data_engineering/
$ echo $STATUS
> READY TO BUILD — currently shipping agents that actually work in production|
🤖 Agentic AI Systems Multi-agent orchestration with LangGraph, stateful workflows, tool use, memory management & complex reasoning chains |
🔍 RAG & Retrieval Vector databases, hybrid search, embedding pipelines, context compression & retrieval-augmented generation at scale |
📊 LLM Evaluation Observability dashboards, tracing, benchmarking, hallucination detection & trust metrics for production LLM systems |
🧠 AI / GenAI / Agents
📊 Data Engineering
☁️ Cloud & Infrastructure
🗄️ Databases & Vector Stores
💻 Languages
🛠️ Frameworks & Dev Tools
|
5 Surprising Truths About Building Real-World AI Agents with LangGraph
|
How Agentic AI Is Redefining How We Work
|
class ManasDani:
def __init__(self):
self.role = "AI Engineer"
self.education = "MS Data Science @ Indiana University Bloomington"
self.currently_building = [
"Production-grade multi-agent systems with LangGraph",
"RAG pipelines with hybrid retrieval & re-ranking",
"LLM evaluation & observability frameworks",
]
self.currently_learning = [
"Agent memory architectures",
"LLM tracing with Langsmith & Arize",
"Frontier model fine-tuning",
]
self.mission = "Make AI measurable, trustworthy, and scalable"
def __repr__(self):
return f"Building agents that think → {self.mission}"
⚡ Built with intent. Deployed with purpose. Powered by caffeine and curiosity.