LinkedIn • Email • rajasimhabolla@gmail.com
- 🛡️ Security Automation: Former CSIRT Intern at Docusign, where I built autonomous AI pipelines that slashed security alert triage times by 94%.
- 💻 Backend Scalability: Designing fault-tolerant microservices, high-throughput prefetch engines, and containerized distributed cloud architectures.
- 🤖 Edge & Agentic AI: Orchestrating multi-agent frameworks (CrewAI) and optimizing LLM streamed inference latency down to 0.3 seconds for edge devices.
- 🏆 Hackathon Competitor: Top 1% out of 50k+ in JPMorgan Chase 'Code for Good' and top 30 in India for ISRO's space tech hackathon.
| Category | Tools & Technologies |
|---|---|
| Languages & Queries | Python, Java, C++, JavaScript, KQL, SQL |
| Security & Automation | Cortex XDR, Microsoft Sentinel (SIEM), CrewAI, n8n, Log Parsing |
| Backend & DevOps | Docker, Kubernetes, Flask, Django, Next.js, REST APIs, Microservices |
| Cloud Infrastructure | AWS, Azure, Google Cloud Platform (GCloud) |
| AI Systems & Safety | Context Normalization, OWASP LLM Mitigation, RAG, PyTorch, TensorFlow |
An autonomous, multi-agent triage pipeline built to eliminate manual analyst bottlenecks.
- The Pipeline: Engineered a parallelized backend that fans out to 7+ enterprise telemetry systems (Sentinel, XDR, VirusTotal) to ingest and normalize 60k+ words of log context into strict JSON payloads.
- The Intelligence: Orchestrated a 3-agent cluster (Endpoint, Intel, Synthesis) running multi-model routing to score alert risks against strict standard operating procedures.
- The Impact: Slashed Mean Time to Triage from 2 hours to under 3 minutes, saving 140+ engineering hours per month with a 94% True Positive accuracy.
- Safety: Layered custom guardrails to harden the agents against prompt injections, data bias, and hallucinations using OWASP guidelines.
A horizontally scalable asynchronous microservices platform designed for dynamic content generation.
- Integrated Gemini and GCloud Voice pipelines to automate news parsing and synthesize high-fidelity audio, dropping processing latency by 50 seconds.
- Implemented a containerized infrastructure using Docker and Kubernetes on hybrid cloud (GCloud/Azure) to handle auto-scaling and traffic load balancing.
- Won the premier idea pitch competition at the BIG G Foundation.
A real-time educational agent designed for constrained hardware environments.
- Fine-tuned a proprietary dataset of 1,000+ instruction-response pairs to boost specialized reasoning accuracy by 9%.
- Optimized streamed inference pipelines using cache-aware token management, driving end-to-end inference latency down to 0.3 seconds for mobile and edge deployment.
- Core Contributions: Active contributor to
Hugging Face Transformersand open-source infrastructure tools. - Data Structures: 1st Place at the BMSCE Byte Battle speed-programming DSA tournament.