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

Hi, I’m Jayesh Locharla, an AI/ML Engineer with 4+ years of experience building and deploying machine learning and Generative AI systems. I specialize in LLMs, NLP, recommendation systems, RAG architectures, and real-time ML pipelines, with strong expertise in Transformers, fine-tuning (LoRA/SFT), and deep learning. I’m proficient in Python, AWS, Spark, Kafka, Docker, and Kubernetes, delivering low-latency AI systems and improving search relevance, CTR, and predictive modeling performance through end-to-end ML system design and production deployment. I’ve led end-to-end ML system design and deployment across Walmart and Goldman Sachs, building transformer-based ranking for e-commerce search, RAG systems, real-time data pipelines, and demand forecasting. I enjoy collaborating across teams and delivering measurable impact through scalable, robust ML solutions.…Hi, I’m Jayesh Locharla, an AI/ML Engineer with 4+ years of experience building and deploying machine learning and Generative AI systems. I specialize in LLMs, NLP, recommendation systems, RAG architectures, and real-time ML pipelines, with strong expertise in Transformers, fine-tuning (LoRA/SFT), and deep learning. I’m proficient in Python, AWS, Spark, Kafka, Docker, and Kubernetes, delivering low-latency AI systems and improving search relevance, CTR, and predictive modeling performance through end-to-end ML system design and production deployment. I’ve led end-to-end ML system design and deployment across Walmart and Goldman Sachs, building transformer-based ranking for e-commerce search, RAG systems, real-time data pipelines, and demand forecasting. I enjoy collaborating across teams and delivering measurable impact through scalable, robust ML solutions.

Jayesh Locharla

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





Hi, I’m Jayesh Locharla, an AI/ML Engineer with 4+ years of experience building and deploying machine learning and Generative AI systems. I specialize in LLMs, NLP, recommendation systems, RAG architectures, and real-time ML pipelines, with strong expertise in Transformers, fine-tuning (LoRA/SFT), and deep learning. I’m proficient in Python, AWS, Spark, Kafka, Docker, and Kubernetes, delivering low-latency AI systems and improving search relevance, CTR, and predictive modeling performance through end-to-end ML system design and production deployment. I’ve led end-to-end ML system design and deployment across Walmart and Goldman Sachs, building transformer-based ranking for e-commerce search, RAG systems, real-time data pipelines, and demand forecasting. I enjoy collaborating across teams and delivering measurable impact through scalable, robust ML solutions.…Hi, I’m Jayesh Locharla, an AI/ML Engineer with 4+ years of experience building and deploying machine learning and Generative AI systems. I specialize in LLMs, NLP, recommendation systems, RAG architectures, and real-time ML pipelines, with strong expertise in Transformers, fine-tuning (LoRA/SFT), and deep learning. I’m proficient in Python, AWS, Spark, Kafka, Docker, and Kubernetes, delivering low-latency AI systems and improving search relevance, CTR, and predictive modeling performance through end-to-end ML system design and production deployment. I’ve led end-to-end ML system design and deployment across Walmart and Goldman Sachs, building transformer-based ranking for e-commerce search, RAG systems, real-time data pipelines, and demand forecasting. I enjoy collaborating across teams and delivering measurable impact through scalable, robust ML solutions.

Available to hire

Hi, I’m Jayesh Locharla, an AI/ML Engineer with 4+ years of experience building and deploying machine learning and Generative AI systems. I specialize in LLMs, NLP, recommendation systems, RAG architectures, and real-time ML pipelines, with strong expertise in Transformers, fine-tuning (LoRA/SFT), and deep learning. I’m proficient in Python, AWS, Spark, Kafka, Docker, and Kubernetes, delivering low-latency AI systems and improving search relevance, CTR, and predictive modeling performance through end-to-end ML system design and production deployment.

I’ve led end-to-end ML system design and deployment across Walmart and Goldman Sachs, building transformer-based ranking for e-commerce search, RAG systems, real-time data pipelines, and demand forecasting. I enjoy collaborating across teams and delivering measurable impact through scalable, robust ML solutions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

AI/ML Engineer at Walmart

September 1, 2024 - Present

Engineered an AI-driven e-commerce search ranking system using Learning-to-Rank models and TensorFlow, modeling a 5M+ user-query product interaction matrix with behavior signals, embeddings, and intent features, improving CTR by 13–19% via relevance optimization. Designed Transformer-based NLP pipelines for semantic understanding of search queries and product catalogs, implemented an Agentic AI-powered RAG system to enhance contextual product discovery and reduce zero-result queries by 20%+. Built end-to-end data pipelines with Spark, SQL, and AWS to process large-scale clickstream data, enabling scalable feature engineering for millions of daily events. Deployed real-time search ranking APIs with FastAPI, Docker, and Kubernetes, achieving sub-150ms latency. Implemented DL-based ranking models (ANN + Transformer) to improve conversion rates by 12–14%. Built a demand forecasting system using LSTM and XGBoost on large-scale SKU data, improving forecast accuracy by 16%.

Machine Learning Scientist at Goldman Sachs

January 1, 2021 - July 1, 2023

Architected a BERT-based earnings intelligence system fine-tuned on 3M+ financial tokens to extract guidance, risks, and forward-looking statements, improving extraction accuracy by 20% over baselines. Developed a Named Entity Recognition pipeline with BiLSTM-CRF and transformer embeddings, achieving F1-score improvement from 0.81 to 0.91. Implemented abstractive summarization (T5/BART-based) to compress long financial reports, reducing analyst reading time by 40–55% while preserving >90% semantic coverage. Built scalable ML pipelines on AWS (SageMaker, S3, EC2) with CI/CD, reducing release time. Applied unsupervised clustering (K-Means, LDA) on millions of documents to improve thematic classification by ~30%. Engineered predictive features across terabytes of data using Logistic Regression and XGBoost to achieve ~60% directional accuracy in post-earnings stock movement predictions.