Nakul Reddy Sarasani

About

Software engineer with JPMorgan Chase experience building a real-time operations console backend on Spring Boot/Kafka/Kubernetes and resolving peak-load latency through profiling, indexing, caching, and async processing. Also built and owned an AI-driven digital-archives metadata pipeline during a master’s at UNT using OCR + LLaMA-based prompting with validation, near-human accuracy, and human-in-the-loop guardrails.

Experience

Education

Key Strengths

• Built real-time backend system (Spring Boot/Kafka/Kubernetes) supporting 10,000+ transactions/users
• Debugged and reduced peak-load latency via profiling, query/index optimization, caching, and async processing
• End-to-end ownership of AI metadata automation pipeline (OCR + LLaMA prompting + validation) achieving near-human accuracy
• Designed reliable multi-step workflow with checkpoints/state machine, retries with exponential backoff, and escalation to manual review
• Strong production engineering practices in Python (typing/mypy, asyncio/Celery, pytest, CI, logging/monitoring)
• Improved slow relational query performance by analyzing query plans and adding targeted indexes (seconds to milliseconds)

Reference Highlights

• Highly effective backend engineer (rated 9/10)
• Strong proficiency in ASP.NET Web API
• Builds well-structured, reusable API controllers and service logic
• Designed and implemented core API integration layer between Angular and ASP.NET
• Strong request/response and error-handling practices
• Improved product discovery and user interaction via LLM assistant
• Reduced navigation friction and time-to-find relevant products
• Disciplined workflow design with clear planning/execution/verification stages
• Uses structured data contracts to improve reliability
• Writes clean, modular Python with separation of concerns
• Strong debugging and issue isolation skills
• Thoughtful schema design with clear entity boundaries and normalization
• Proactively enforces data integrity (e.g., preventing duplicates with constraints)
• Takes ownership in ambiguous 0→1 work; doesn’t wait for detailed requirements
• Moves work forward with iterative validation
• Incorporates feedback early through short loops; avoids late-stage rework
• Makes pragmatic speed/quality tradeoffs via phased scope and prioritization
• Highly effective backend engineer (9.5/10)
• Consistently reliable end-to-end delivery
• Strong at clarifying requirements and expectations
• Solid validation and structured error handling
• Collaborates well with frontend and DevOps
• Good engineering judgment on scalability and operational readiness
• Improved system reliability and reduced production issues
• Led an LLM-driven metadata enhancement pipeline
• Strong prompt engineering with structured prompting
• Built evaluation-driven systems; minimizes hallucinations
• Near-human metadata extraction performance (F1: 0.93/0.89/0.90)
• Delivered measurable business impact (reduced manual effort; helped secure funding)
• Strong at designing reliable multi-step workflows in distributed systems
• Implemented idempotency, health checks, and enhanced monitoring
• Reduced production defects by 30%
• Strong Python code clarity, modularity, and maintainability
• Effective debugging via logging and validation
• Solid relational database skills (schema, migrations, indexing, query optimization)
• Strong ownership in ambiguous/early-stage work
• Methodical, adaptable approach; explicit assumptions and early validation
• Balances speed, quality, and product outcomes with sound tradeoffs
• Prioritizes long-term quality and reliability over short-term speed
• Strong fundamentals and clarity in backend concepts
• Able to simplify complex topics effectively
• Highly effective contributor
• Focus on reliability and structured approaches
• Improved LLM output consistency and accuracy to near-human levels
• Clean, modular code
• Methodical debugging using metrics and validation
• Good judgment in relational data design and integrity
• Takes complete ownership, even with unclear requirements
• Proactive at unblocking obstacles and coordinating across groups
• Calm demeanor
• Learns very fast
• Strong team player
• Prioritizes long-term quality and reliability over speed when needed