GenAI Engineer building production-grade LLM systems, agentic workflows, and AI infrastructure automation.
Currently working on large-scale telecom AI systems and cloud optimization platforms. I focus on designing reliable, observable, and cost-efficient AI systems that operate beyond notebooks.
- 🤖 Agentic AI systems using LangGraph, MCP, and multi-agent orchestration
- 🔎 Graph + Hybrid RAG pipelines for high-precision retrieval
- 📊 LLMOps & observability with LangFuse, MLflow, and Triton
- ☁️ Cloud cost optimization & GPU utilization analytics
- ⚡ Low-latency inference systems with quantized and fine-tuned models
- Automated 60% of telecom network triage using multi-agent RAG
- Built Kubernetes AI debugger reducing infra triage time by 60%
- Delivered $1.5M+ annual cloud savings through GPU observability
- Optimized LLM inference latency by 35% via LoRA + Triton deployment
- Improved outage detection across 5K+ daily alarms using predictive modeling
LLM & Agents
LangGraph • LangChain • MCP • CrewAI • Autogen • LlamaIndex
RAG & Retrieval
Hybrid RAG • Graph RAG • Neo4j • Elastic Vector DB • pgvector • Cohere Rerank
LLMOps & Infra
Ray • ONNX • NVIDIA Triton • MLflow • LangFuse • Kubeflow • Run:AI
Data Engineering
Spark • Airflow • Kafka • Snowflake • TimescaleDB • dbt
Backend & Cloud
Python • FastAPI • Java • AWS • GCP • Kubernetes • Terraform • Docker
ApplyAI
LLM-powered job automation system with async processing and multi-model routing.
Stack: FastAPI • Next.js • Celery • Redis • Vector embeddings
SpringCommerce
Microservices e-commerce platform with polyglot persistence and observability stack.
AtmoFlow
Batch + streaming data pipelines using Spark, Airflow, and GCP.
- Production AI systems
- Agent orchestration standards (MCP / A2A)
- Scalable RAG architectures
- Efficient LLM inference
- AI infra reliability
- LinkedIn: https://linkedin.com/in/rahul-reddy-t
- Portfolio: https://rahultalatala.netlify.app
- GitHub: https://github.com/rahult18