About
Builder Profile
AI engineer building LLM workflows, ML services, backend APIs, and dashboards that make intelligent systems usable in the real world.
Strong in Python, systems thinking, applied research, and fast product execution. Currently focused on AI governance, cost analytics, and agentic automation.
Education
B.Tech in Artificial Intelligence and Data Sciences
KL University Hyderabad
Aug. 2022 – May 2026 • GPA: 9.38
Board of Intermediate Education
Sri Chaitanya Junior College
Aug. 2020 – May 2022 • 91.9%
Board of Secondary Education
Sri Chaitanya School SP Nagar
Aug. 2019 – June 2020 • GPA: 10.0
Skills
Languages
Core programming languages for development and data manipulation
AI/ML Tools
Specialized tools and models for AI/ML development and deployment
Frameworks
Development frameworks for building scalable applications
Cloud & DevOps
Cloud services, containerization, and API development
IDEs & Tools
Development environments and version control
Experience
Associate AI Python Engineer
Techolution
Enterprise AI governance platform — cost intelligence, root-cause analysis, and cloud automation. Promoted from AI Engineering Intern.
- Built the flagship AI Governance platform’s L4 Entity Intelligence layer, unifying cost, usage, and movement analytics across 11 entity scopes including Firestore, Compute Engine, AlloyDB, Cloud Logging, Pub/Sub, and agent-assistant relationships
- Designed movement-detection logic (spike, drop, fast, slow, new, dormant) backed by 3/7/30-day moving averages and working-day-aware historical comparisons to surface the resources that actually need investigation
- Engineered a generic multi-source AI root-cause-analysis agent that pulls evidence from AlloyDB, Cloud Logging, and PromQL metrics, and auto-triggers on significant cost movements
- Added hallucination-prevention guardrails — schema and column validation, contract checks — and a structured Incident / Root Cause / Next Steps report format with evidence-ranked contributors
- Optimized historical data backfills, validating a 5,674-row, 219-call run with zero failures while cutting batch-processing time roughly 13x
- Automated daily intelligence refreshes with Cloud Scheduler and a 24h/48h reconciliation window to account for delayed cloud billing data
- Shipped supporting platform features: Triage AI alert ingestion and classification, RBAC-backed Smart AI Prompt Policies with audit trails, LLM and cloud cost analytics, and cloud utilization/hibernation automation
Data Science Intern
PiResearch Labs
Operational data systems, backend APIs, and workflow automation
- Designed and implemented the database schema that supported company-wide operations with a focus on scalability and reliability
- Built and optimized REST API endpoints to support core application workflows and improve system performance
- Automated reconciliation and merchant background-check workflows to reduce manual effort and improve operational efficiency
Contractor - Agentic AI
Hexagon R&D India
Agentic root-cause analysis for enterprise database operations
- Conceptualized and developed an Agentic AI prototype with autonomous root cause analysis for databases, improving HxGN NetWorks issue resolution speed by 25% and cutting average resolution time by 40%
- Designed AI agents using Azure OpenAI models with MagenticOne to autonomously analyze database and service logs, reducing resolution time by 40%; benchmarked against state-of-the-art Agentic frameworks like AutoGen and CrewAI
- Conducted extensive experimentation with REST APIs and Ollama models, improving system scalability by 20% and reducing cloud computing costs compared to generic GPT-based solutions
AI Intern
Arthink.ai
Predictive ML, anomaly detection, and LLM-assisted monitoring
- Developed and deployed regression models on NASA CMAPSS dataset for predicting engine Remaining Useful Life (RUL), achieving 86.24% accuracy using TensorFlow-based architectures
- Engineered a full-stack Flask application integrating predictive ML models, clustering techniques, Granger Causality for anomaly detection, and Langchain-based LLM agents, resulting in a production-ready AI monitoring tool
- Fine-tuned GPT-2 using Hugging Face Transformers and LoRA on domain-specific text data, boosting downstream NLP task performance by 30% compared to baseline pre-trained models
Projects
Publications
Enhancing Remaining Useful Life Prediction
A Comparative Study of Classical Machine Learning and Generative AI
Springer CCIS Series • ThinkAI'24 Conference
Certifications
AWS Certified Cloud Practitioner
Amazon Web Services (AWS)
ID bd40fb4be29e4f4bb03a6c01c4af72be
Salesforce Certified Agentforce Specialist
Salesforce
ID 7211045
Microsoft Certified: Azure Fundamentals
Microsoft
ID 6AC6F1A92FF0E07C
Blog
Notes on applied AI, machine learning systems, and the engineering decisions behind real projects.
Get In Touch
Contact Channels
Start a Thread
✓ Available for Opportunities
Open to full-time positions, freelance projects, and collaboration opportunities in AI/ML, full-stack development, and cloud technologies.