Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Harinath Chakali, an AI/ML Engineer with 3+ years of experience building machine learning and Generative AI systems across enterprise infrastructure and large-scale consumer platforms. I have designed Retrieval-Augmented Generation (RAG) pipelines, LLM evaluation frameworks, and production ML services using Python, PyTorch, LangChain, Azure OpenAI, and Kubernetes. I collaborate with data scientists, reliability engineers, and platform teams to deliver scalable AI solutions that integrate predictive modeling, distributed data pipelines, and production-grade LLM applications for operational intelligence.…I'm Harinath Chakali, an AI/ML Engineer with 3+ years of experience building machine learning and Generative AI systems across enterprise infrastructure and large-scale consumer platforms. I have designed Retrieval-Augmented Generation (RAG) pipelines, LLM evaluation frameworks, and production ML services using Python, PyTorch, LangChain, Azure OpenAI, and Kubernetes. I collaborate with data scientists, reliability engineers, and platform teams to deliver scalable AI solutions that integrate predictive modeling, distributed data pipelines, and production-grade LLM applications for operational intelligence.

HARINATH CHAKALI

AI Engineer, Data Scientist, Full Stack Developer, +2





I'm Harinath Chakali, an AI/ML Engineer with 3+ years of experience building machine learning and Generative AI systems across enterprise infrastructure and large-scale consumer platforms. I have designed Retrieval-Augmented Generation (RAG) pipelines, LLM evaluation frameworks, and production ML services using Python, PyTorch, LangChain, Azure OpenAI, and Kubernetes. I collaborate with data scientists, reliability engineers, and platform teams to deliver scalable AI solutions that integrate predictive modeling, distributed data pipelines, and production-grade LLM applications for operational intelligence.…I'm Harinath Chakali, an AI/ML Engineer with 3+ years of experience building machine learning and Generative AI systems across enterprise infrastructure and large-scale consumer platforms. I have designed Retrieval-Augmented Generation (RAG) pipelines, LLM evaluation frameworks, and production ML services using Python, PyTorch, LangChain, Azure OpenAI, and Kubernetes. I collaborate with data scientists, reliability engineers, and platform teams to deliver scalable AI solutions that integrate predictive modeling, distributed data pipelines, and production-grade LLM applications for operational intelligence.

Available to hire

I’m Harinath Chakali, an AI/ML Engineer with 3+ years of experience building machine learning and Generative AI systems across enterprise infrastructure and large-scale consumer platforms. I have designed Retrieval-Augmented Generation (RAG) pipelines, LLM evaluation frameworks, and production ML services using Python, PyTorch, LangChain, Azure OpenAI, and Kubernetes. I collaborate with data scientists, reliability engineers, and platform teams to deliver scalable AI solutions that integrate predictive modeling, distributed data pipelines, and production-grade LLM applications for operational intelligence.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

AI Engineer at Uber

August 1, 2025 - Present

Designed a Retrieval-Augmented Generation (RAG) pipeline using LangChain, Azure OpenAI, and FAISS indexing 320K+ ride operations documents to automate responses for 160K monthly internal support queries. Developed multi-agent diagnostic workflows using LangGraph and Hugging Face Transformers for ride-matching incident analysis, integrating NVIDIA NeMo LLM guardrails that blocked 6K unsafe prompts weekly. Established an LLM evaluation framework using RAGAS, curated golden datasets, and LLM-as-a-Judge scoring, testing 30 prompt configurations and identifying retrieval improvements that lowered average response latency to 2.4 seconds. Optimized model inference services using vLLM with Dockerized deployment on Kubernetes, supporting 95K+ daily LLM inference requests for internal automation tools while reducing average GPU memory usage by 6GB per container. Automated model lifecycle management using MLflow tracking, FastAPI inference endpoints, and GitHub Actions CI/CD, enabling reproducibl

Machine Learning Engineer at Dell Technologies

January 1, 2021 - July 1, 2023

Engineered predictive maintenance models using Python, Scikit-learn, and PySpark analyzing 38M+ hardware telemetry records from enterprise storage systems, generating automated alerts that prevented 2,900 device failures annually. Built distributed data pipelines using Apache Spark and Apache Airflow to ingest and validate telemetry streams from 10 global data centers, standardizing feature engineering workflows used across multiple anomaly detection models. Refined anomaly detection training workflows using TensorFlow with hyperparameter tuning tracked in MLflow, reducing model training cycles from 10.6 hours to 3.8 hours while improving stability of storage performance predictions. Implemented streaming ingestion pipelines using Apache Kafka and Spark Structured Streaming, enabling near-real-time feature updates for predictive maintenance models processing 120K telemetry events per minute. Deployed containerized ML inference services using Docker and Kubernetes on AWS, exposing REST