Archive · 5 Nov 2025
Saikrishna Garlapati Engineers Cloud-Native AI for Fraud Prevention in Banking
Imagine banks are like castles protecting treasure from sneaky thieves (fraudsters).
Old castles used simple fences and guards (rule-based systems) that weren't always good enough.
Now, smart engineers like Saikrishna Garlapati are building super-modern castles with high-tech tools.
They use special computer brains (AI) and flexible building blocks (cloud-native tech) that can spot thieves instantly, even when lots of people are visiting (high-throughput transactions).
These new systems can fix problems really fast, like sending new magic keys (card replacements) right away.
They also follow strict rules to keep everything safe and reduce costs for the castle owners.
This helps keep everyone's money safer and makes customers happier because their treasure is better protected.
The banking industry is transitioning from legacy systems to cloud-native and AI-powered solutions for real-time fraud detection and remediation.
Saikrishna Garlapati specializes in designing these advanced fraud systems, utilizing technologies like AWS Lambda and SageMaker.
His work has successfully processed over $2 billion in fraud claims, resulting in significant cost reductions and improved transaction speeds.
Garlapati has engineered systems for real-time card replacement and credit issuance, enhancing customer satisfaction and fraud resilience.
Key innovations include overcoming serverless cold start latency and using AI for real-time fraud triage, emphasizing regulatory compliance and zero-trust security.
- Who
- Saikrishna Garlapati, an engineer with expertise in fraud remediation systems for major U.S. banks.
- What
- Developing and implementing cloud-native, AI-powered solutions to optimize fraud detection and remediation, focusing on real-time processing, scalability, and regulatory compliance.
- When
- The shift is happening currently, with publications and work spanning up to March 2025.
- Where
- Primarily within the U.S. banking sector.
- Why
- To combat billions of dollars at risk from fraud, maintain customer trust, reduce operational costs, and meet stringent regulatory standards.
Legacy Systems
Modern Cloud-Native & AI Solutions
Fraud Detection Method
Legacy Systems
Relies on rigid, rule-based systems.
Modern Cloud-Native & AI Solutions
Utilizes intelligent, event-driven pipelines and AI-based detection.
Scalability & Speed
Legacy Systems
Struggles with high-throughput transactions and real-time needs.
Modern Cloud-Native & AI Solutions
Designed for high-throughput processing and real-time identification, mitigation, and remediation.
Operational Expenses
Legacy Systems
Higher infrastructure costs.
Modern Cloud-Native & AI Solutions
Achieves quantifiable results like reduced infrastructure expenses (e.g., 19% decrease).
Customer Impact
Legacy Systems
Longer remediation cycles and potential for duplicate credits.
Modern Cloud-Native & AI Solutions
Enables real-time card replacement and credit issuance, improving customer satisfaction.
- Lead Engineer
- Saikrishna Garlapati
- Focus Area
- Fraud detection and remediation in banking
- Key Technologies
- Cloud-native, AI, AWS Lambda, SageMaker, serverless frameworks
- Fraud Claim Processing
- Over $2 billion processed
- Operational Improvements
- 19% decrease in infrastructure expense, 0.88 F1-score fraud model, 300ms transaction processing latency
- Regulatory Compliance
- PCI-DSS monitoring, zero-trust security principles