Sudhamsu Bairisetti

Machine Learning Engineer

Cincinnati, Ohio, United States4 yrs 8 mos experience

AI ML PractitionerAI Enabled

Key Highlights

Improved fraud detection and recommendation systems significantly.
Built scalable ML pipelines with measurable impact on efficiency.
Led Generative AI initiatives with substantial cost reductions.

Stackforce AI infers this person is a Machine Learning Engineer with expertise in AI solutions for E-commerce and Financial Services.

Contact

Skills

Core Skills

Machine LearningGenerative AiData ScienceData EngineeringSoftware DevelopmentData Visualization

Other Skills

NLPtransformer-based modelsHugging FaceOpenAI APIsAirflowMLflowFastAPIDockerKubernetesvector databasesFAISSPineconehyperparameter optimizationOptunaRay Tune

About

AI/ML Engineer (Machine Learning Engineer) with 4+ years of experience designing, building, and deploying scalable machine learning, deep learning and GenAI solutions across financial services and enterprise AI platforms. Strong expertise in Python, TensorFlow, PyTorch, NLP, Computer Vision, Generative AI and MLOps. Experienced in building data pipelines with Airflow, Kafka, and ETL frameworks, and deploying models using AWS SageMaker, GCP Vertex AI. Proven ability to improve fraud detection, recommendation systems, and decision automation through production grade AI systems. Experienced in building end to end ML pipelines across cloud and on premise environments with measurable impact on accuracy, cost reduction, and deployment efficiency.

Experience

4 yrs 8 mos

Total Experience

3 yrs 4 mos

Average Tenure

1 yr 4 mos

Current Experience

Openai

Machine Learning Engineer

Jan 2025 – Present · 1 yr 4 mos · United States

Developed and fine-tuned NLP and transformer-based models using Hugging Face and OpenAI APIs, improving summarization and classification accuracy by 30%+.
Built end-to-end ML pipelines using Airflow and MLflow, automating ingestion, training, validation, deployment, and monitoring — reducing manual effort by 40%+.
Deployed scalable real-time inference services using FastAPI, Docker, and Kubernetes supporting 1M+ API calls per day with 99.9% uptime.
Integrated vector databases (FAISS, Pinecone) for semantic search and retrieval-augmented generation (RAG), improving retrieval speed by 40%.
Applied hyperparameter optimization using Optuna and Ray Tune, boosting F1 scores by up to 18%.
Built monitoring dashboards with Prometheus and Grafana to detect model drift, latency, and performance degradation, reducing downtime by 60%.
Created automated retraining workflows triggered by data drift detection in AWS SageMaker, ensuring models remained production-ready.
Worked on Generative AI chatbots with transformer-based embeddings, improving contextual accuracy by 25%.
Designed automated retraining workflows triggered by drift detection using AWS SageMaker.
Led Generative AI initiatives for document intelligence, chatbot systems, and automated financial insights generation, reducing manual review efforts by 40%.
Implemented explainable AI techniques (SHAP, LIME) to enhance transparency and regulatory compliance.

NLPtransformer-based modelsHugging FaceOpenAI APIsAirflowMLflow+14

Flipkart

Data Scientist (Machine Learning)

Jul 2020 – Nov 2023 · 3 yrs 4 mos · India

Built recommendation systems using collaborative filtering and deep learning, improving engagement metrics and conversion rates.
Processed 10TB+ structured and unstructured datasets using PySpark, Hive, and Hadoop reducing ETL runtime by 35% and infrastructure costs by 20%.
Developed fraud detection systems using Isolation Forests, Autoencoders, XGBoost, and ensemble techniques reducing false positives by 20%+.
Engineered advanced feature pipelines using SQL, Spark, and embedding techniques (PCA, t-SNE) reducing training time by 15%.
Built real-time ingestion pipelines using Kafka and AWS Glue reducing latency by 40%.
Deployed and automated model retraining using AWS SageMaker, Docker, Kubernetes, and CI/CD pipelines cutting deployment cycle time by 30%.
Developed REST APIs using Flask to integrate ML services into enterprise platforms serving millions of users.
Implemented computer vision models (YOLO, CNNs, OpenCV) for product classification achieving 95%+ accuracy.
Monitored production models using AUC-PR, PSI, CSI, and Gini metrics to ensure model stability and compliance.
Created executive dashboards using Tableau and Power BI accelerating decision-making by 35%.