
Prithviraj G
AI Backend Engineer
Skills

Bekijk mijn diensten

Werkervaring
AI Engineer
Coditas • Fulltime
Jul 2025 - Present • 10 mos
Architected and delivered AI-first backend services for an enterprise GRC platform, enabling teams to replace hours of manual work with automated workflows. Built an English-to-Business-Rules conversion service using LangGraph (multi-node pipeline) + RAG (Weaviate), cutting BR authoring effort by ~40%. Designed a CEC-file upgrade automation system with an async RabbitMQ consumer, Redis Pub/Sub progress streaming, and a WebSocket-based real-time UI — handling 10,000+ objects per file. Engineered MCP (Model Context Protocol) tools integrated with Windsurf Workflows, reducing Change History Report registration from 3–4 hours to under 7 minutes per form. Automated Business API registration pipeline by wrapping instance CLI commands as REST APIs and orchestrating them via custom MCP tools. Identified and resolved a critical multi-tenant DB connection bug in a production MCP server, preventing data crosscontamination across customer instances. Designed Redis-based distributed locking and RabbitMQ-driven bulk-save architecture respecting strict third-party API concurrency limits (parallel form locks, 100-BR batches, async job tracking).
Robotics Software Engineer
FlytBase • Fulltime
Oct 2023 - Apr 2025 • 1 yr 6 mos
Developed an object detection and tracking system utilizing the YOLOv5 model, OpenCV tracking algorithms, and a PID controller, integrated with ROS to enable autonomous tracking with DJI drones, using C++. Architected and developed Python-based cloud microservice systems to manage streaming protocols, airspace monitoring, and hardware abstraction, integrating with DJI Cloud SDK and ROS2 frameworks. Enhanced performance of FlytBase Android Application using Kotlin by fixing the crashing issues. Thus improving the overall performance by 25%. Increased user attach rate (by 40%) to legacy systems by developing features like Cloud Media Sync and 4G Support for scalable drone control via MQTT and DJI Mobile SDK over Android Application.
Python Developer
AlgoBulls • Fulltime
Dec 2022 - Oct 2023 • 10 mos
Built and optimised algorithmic trading strategies using TA-Lib and integrating different brokers in our system. Built data visualisation tools, to analyse and generate reports of trades. Integrated Prompt Engineered Chat-GPT APIs to build an interface for users to generate their own strategy using pyalgotrading libraries.