I am an AI/ML Engineer with around five years of experience designing, developing, and deploying production Machine Learning and Generative AI systems across financial services and intelligence-driven applications. I specialize in LLMs, Retrieval-Augmented Generation (RAG), Deep Learning, NLP, fraud detection, and Agentic AI using Python, PyTorch, TensorFlow, SQL, Spark, AWS, and Docker. I have built end-to-end AI solutions spanning data engineering, model development, LLM fine-tuning, vector search, evaluation frameworks, and inference systems using LangChain, LangGraph, Hugging Face Transformers, and modern LLMOps practices. I have delivered platforms and solutions in fast-paced financial services environments, collaborating with analysts and engineers to ship reliable, observable, and scalable AI systems. My work emphasizes accuracy, guardrails to mitigate hallucinations, and efficient deployment to production through robust MLOps practices.

Keerthana Venkata

I am an AI/ML Engineer with around five years of experience designing, developing, and deploying production Machine Learning and Generative AI systems across financial services and intelligence-driven applications. I specialize in LLMs, Retrieval-Augmented Generation (RAG), Deep Learning, NLP, fraud detection, and Agentic AI using Python, PyTorch, TensorFlow, SQL, Spark, AWS, and Docker. I have built end-to-end AI solutions spanning data engineering, model development, LLM fine-tuning, vector search, evaluation frameworks, and inference systems using LangChain, LangGraph, Hugging Face Transformers, and modern LLMOps practices. I have delivered platforms and solutions in fast-paced financial services environments, collaborating with analysts and engineers to ship reliable, observable, and scalable AI systems. My work emphasizes accuracy, guardrails to mitigate hallucinations, and efficient deployment to production through robust MLOps practices.

Available to hire

I am an AI/ML Engineer with around five years of experience designing, developing, and deploying production Machine Learning and Generative AI systems across financial services and intelligence-driven applications. I specialize in LLMs, Retrieval-Augmented Generation (RAG), Deep Learning, NLP, fraud detection, and Agentic AI using Python, PyTorch, TensorFlow, SQL, Spark, AWS, and Docker. I have built end-to-end AI solutions spanning data engineering, model development, LLM fine-tuning, vector search, evaluation frameworks, and inference systems using LangChain, LangGraph, Hugging Face Transformers, and modern LLMOps practices.

I have delivered platforms and solutions in fast-paced financial services environments, collaborating with analysts and engineers to ship reliable, observable, and scalable AI systems. My work emphasizes accuracy, guardrails to mitigate hallucinations, and efficient deployment to production through robust MLOps practices.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert

Work Experience

AI/ML Engineer at Goldman Sachs
September 1, 2025 - Present
Built an end-to-end RAG and LLMOps platform for financial intelligence using Transformer-based LLMs and LangChain orchestration, enabling citation-grounded question answering across 250K+ SEC filings and earnings reports with retrieval grounding and guardrails. Architected an embedding and semantic retrieval pipeline using Sentence Transformers, OpenAI embeddings, FAISS indexing, and context engineering, supporting semantic search across 2M+ financial document chunks for analyst research workflows. Fine-tuned domain-adapted LLMs using LoRA and PEFT techniques, incorporating human feedback loops and evaluation workflows with RAGAS and TruLens to improve response quality and model observability. Developed a scalable FastAPI inference platform integrating prompt versioning, model routing, tool calling, function calling, and agent orchestration workflows, supporting 80+ concurrent inference requests with sub-second average latency during load testing. Implemented multi-source ingestion pip
Machine Learning Engineer at JPMorgan Chase & Co.
August 1, 2021 - December 1, 2024
Engineered a real-time fraud detection platform using Python, Scikit-learn, XGBoost, NumPy, and Pandas, processing over 8-10M transaction records to identify fraudulent payment activities across credit card and digital banking channels. Operationalized and optimized more than 500 behavioral, transactional, and merchant-level features using SQL, PySpark, and feature engineering methodologies, improving risk scoring accuracy across high-volume financial transaction datasets. Formulated and trained machine learning and deep learning models using XGBoost, Neural Networks, PyTorch, and Hugging Face Transformers for fraud classification, anomaly detection, and risk assessment across millions of daily transaction events. Automated low-latency streaming pipelines using Apache Kafka, Apache Spark, and Airflow, enabling near real-time processing and monitoring of over 20K transactions per minute for fraud detection workflows. Executed scalable ingestion and transformation pipelines using SQL, Py

Education

Master of Science in Engineering Data Science at University of Houston, Houston, TX
January 11, 2030 - June 29, 2026

Qualifications

Add your qualifications or awards here.

Industry Experience

Financial Services, Software & Internet, Professional Services