Looking for a job

Sai Kamal Kadarla

@saikamalkadarla

AI Engineer building production-grade LLM, RAG, and agentic systems with scalable pipelines and cloud-native architectures.

India

Message

What I'm looking for

I’m looking for a team where I can build production-grade LLM/RAG and agentic systems end-to-end—designing scalable pipelines, retrieval and evaluation, and robust backend services on AWS/GCP/Azure with real-world impact.

I’m an AI Engineer with 5+ years of experience building production-grade AI systems—especially LLM applications, RAG platforms, and agentic workflows. At Ernst & Young, I engineered a high-performance chat platform with persistent conversation history, designed optimized PostgreSQL schemas for embeddings and evaluation telemetry, and delivered distributed RAG pipelines that dramatically cut document onboarding time.

I also focus on reliability and measurable quality: I built hybrid retrieval (vector similarity + BM25), an automated RAG evaluation framework using RAGAS, and human-in-the-loop Text-to-SQL workflows to validate intent, schema selection, and generated SQL. Previously at Modak Analytics, I delivered RAG for multi-document extraction, fine-tuned LLMs with QLoRA, and deployed computer vision models (MobileNetV2, YOLOv8) with strong accuracy—alongside scalable backend messaging frameworks and data pipelines across cloud environments.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

Ernst & Young

Nov 2024 - Present (1 year 8 months)

Engineered a production chat platform and enterprise RAG pipelines, including PostgreSQL-backed metadata/session management and distributed Kafka-based document processing. Built agentic LangGraph workflows and an automated RAG evaluation framework to monitor retrieval and response quality.

Python fastAPI PostgreSQL RAG OpenAI Text Embedding 3 Large DataStax Astra Kafka Embeddings BM25 LangGraph LLM As A Judge Top K Metrics Cosine Similarity Ragas Text To SQL Airflow Pub Sub BigQuery Apache Spark GCP

LLM Engineer

Modak Analytics

Apr 2024 - Nov 2024 (7 months)

Built a multi-document RAG application for Ryobi and fine-tuned text generation and vision models for entity extraction, NER, and classification tasks. Deployed fine-tuned models using Hugging Face TGI and Edge/YOLO workflows to improve scalable inference and detection accuracy.

RAG QLoRA Mistral LLaMA Huggingface TGI MobileNetV2 YOLOv8 NER Pest and Crop Disease Classification

Software Development Engineer

Modak Analytics

Apr 2021 - Apr 2024 (3 years)

Developed Java-based messaging and orchestration components for product pipelines, and implemented data migration, metadata crawling, and notification features. Built and optimized ETL/data workflows using SQL, Scala/Spark, JSON, and cloud integrations across AWS, Azure, and Google Cloud.