work history
Career
Machine Learning Engineer (Intern)
Skynyte
January 2026 - June 2026
- Built the AI/LLM integration layer of a production web intelligence framework using LangChain, integrating 6 model providers (OpenAI GPT-4, Gemini, Mistral, Ollama, AWS Bedrock) via a unified provider-agnostic interface.
- Reduced token consumption through HTML minification and tiktoken-based context window management; enforced structured outputs via Pydantic v2 schema validation with retry logic.
- Designed prompt engineering strategies to guide LLM extraction accuracy across multi-step LangChain graph pipelines, optimising instructions for structured JSON output generation.
Intern [Personal Project]
Nimble Vision
February 2025
- Trained a custom YOLOv8 model to detect pets, humans, and trucks for use in surveillance camera systems, achieving 72% accuracy.
- Leveraged Qualcomm AI Hub to compile, optimize, and deploy the model efficiently on Samsung S22 series devices.
- Designed the solution with real-time performance and mobile deployment in mind for smart surveillance applications.
Intern [Team Project]
Nimble Vision
July 2024 - August 2024
- Backend integration for IOT devices, achieved seamless improved system stability with LORAWAN network servers.
- Architected and implemented a scalable solution using Django, HTML, CSS, JavaScript.
Intern [Team Project]
Oil and Natural Gas Corporation (ONGC)
May 2024 - July 2024
- Automated payroll process for ONGC, significantly reduced manual effort, minimized errors, and enhanced timely, accurate payroll delivery.
- Designed and implemented a streamlined system using SAP ABAP to automate data retrieval, processing, and pay slip pdf generation.
