.png){kind=logo}
Manish Vemula
Mid-level Machine Learning Engineer specializing in real-time pipelines and NLP/GenAI
Connect with Manish
Manish already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
ML/MLOps practitioner from Discover Financial who built and deployed a real-time AI fraud detection platform (LSTM + VAE) on AWS SageMaker with Docker/FastAPI and Jenkins-driven CI/CD. Demonstrated measurable impact (30% accuracy lift, 25% fewer false alerts) and deep expertise in class-imbalance mitigation, drift monitoring, and orchestration (Airflow/Kubeflow), plus strong stakeholder adoption via Power BI dashboards for fraud/compliance teams.
Experience
Education
Key Strengths
- Deployed production fraud detection system at Discover using LSTM + VAE (real-time)
- Improved detection accuracy by 30% and reduced false alerts by 25%
- Handled extreme class imbalance (<2% fraud) using SMOTE/undersampling + weighted/focal loss; improved fraud recall by 20%+ without increasing false positives
- Built MLOps deployment and retraining automation (Docker, FastAPI, SageMaker, Jenkins CI/CD)
- Orchestrated end-to-end retraining with Airflow DAGs; reduced retraining cycle from 2 days to a few hours
- Strong reliability practices: modular pipelines, validation checkpoints, unit tests, A/B tests, shadow deployments, drift monitoring with automated retraining triggers
- Effective communication with non-technical stakeholders; built Power BI dashboard translating model performance into business metrics
Discover more candidates like Manish
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent