Rajat Mishra

Skydio

San Mateo, CA, USA — Software Engineer Intern (Jun 2025 – Dec 2025)

• Built a real-time subject tracking system for the Skydio X-10 autonomous drone using Go, Python, and GPT-4, achieving 95% accuracy at 60 FPS with multi-threaded processing and natural-language prompts.
• Designed and delivered an LLM-powered compensation tool in Python Streamlit using LangChain and FAISS to match job descriptions across 10K+ roles (200K+ records), delivering salary insights with feedback-driven refinement.
• Shipped a production offboarding platform (Node.js, React, Typescript, PostgreSQL on Kubernetes) automating workflows via DocuSign and Workday API integrations, cutting document creation time by 90%.

Rigi

Bengaluru, India — Software Engineer II (Feb 2023 – Jun 2024)

• Led development of an in-house HLS transcoding pipeline with FFmpeg on AWS EC2, processing 10K+ videos monthly at 2x faster speeds while cutting costs by 40% and eliminating dual-format overhead.
• Architected and owned a Redis-based watch progress system handling 5M+ concurrent devices and 50M+ daily events, enabling seamless cross-device video resumption with sub-second latency while preventing database overload.
• Redesigned the invoicing system with AWS Lambda serverless architecture, cutting cloud costs by 80% while auto-scaling to 2M+ weekly requests with zero infrastructure management.
• Built automated CI/CD pipelines with testing and deployment for containerized microservices using GitHub Actions, AWS CodeDeploy, Jest, and Kubernetes, reducing deployment time by 35%.

Cutshort

Pune, India — Backend Developer (Aug 2022 – Jan 2023)

• Transformed a legacy mailing system into Java Spring Boot microservices with independent scaling and deployment, achieving 99.9% uptime while processing 1M+ weekly requests at sub-100ms latency.
• Implemented Terraform-based Infrastructure as Code for AWS (EC2, Lambda, RDS), enabling reproducible deployments across staging, and production environments and reducing provisioning errors by 80%.
• Optimized MongoDB query performance by 40% by consolidating multiple database calls into single aggregation pipelines with strategic indexing, dramatically reducing latency for complex queries.

Mygate

Bengaluru, India — Software Developer (Jul 2020 – Aug 2022)

• Developed scalable data pipelines using Node.js (Javascript), Kafka, AWS Kinesis, and Glue to process 30M+ daily events in real-time, enabling analytics dashboards and data-driven decision-making for 2M+ residents.
• Reduced AWS S3 storage costs by 80% by converting JSON data to compressed Apache Parquet format, enhancing query performance with AWS Athena by 10%.
• Deployed monitoring tools like Prometheus and Grafana on AWS Linux using Docker, improving system observability and reducing incident resolution time by over 40%.