Kiran is a Machine Learning Engineer with 5. 5 years of experience specializing in AI and ML technologies.
Led the execution of an AI-powered voice ordering system for restaurants, processing orders in Arabic.
Developed an automated system for generating and applying to job pitches on Upwork using Langchain and LLMs.
Designed and implemented real-time speech-to-text transcription and conversational AI responses for an audio assistant.
Demonstrated new video content generation with voice cloning and lip-syncing in an AI-Video-Message project.
Successfully executed an AI-powered voice ordering system for restaurants, enhancing efficiency.
Automated job pitch generation and application for a freelance platform, increasing application speed.
Developed an audio assistant combining real-time transcription, conversational AI, and speech synthesis, improving user interaction.
Led the project for an AI-Video-Message POC, achieving natural and precise lip movements using Wav2Lip.
Overview: Developed an automated system for generating and applying to job pitches on Upwork. Responsibilities: Generated project information using Langchain and a large language model. Integrated Upwork's RSS Feed and utilized Sentence Transformer for job identification. Automated the job application process using Selenium.
Key outcomes:
Automated job pitch generation and application for a freelance platform.
Overview: Led the successful execution of an AI-powered voice ordering system for restaurants. Responsibilities: Managed the project lifecycle, ensuring timely delivery. Engineered voice processing capabilities using GPT 3.5 and other technologies. Integrated voice recognition and text-to-speech technologies.
Key outcomes:
Successfully executed an AI-powered voice ordering system.
Overview: Developed a real-time audio assistant with three core functionalities. Responsibilities: Designed and implemented real-time speech-to-text transcription using Deepgram and Whisper AI. Integrated conversational AI for contextually appropriate responses using OpenAI GPT-4.
Key outcomes:
Developed an audio assistant combining real-time transcription, conversational AI, and speech synthesis.
Overview: Demonstrated the feasibility of generating new video content with an original person's voice and accurate lip-syncing. Responsibilities: Led the project from conception to completion, ensuring timely delivery of the proof of concept (POC). Used Whisper model to generate video transcripts.
Key outcomes:
Led the project from conception to completion for a video content generation POC.
Overview: Developed a script to collect and analyze images for a Guinness World Record tree planting campaign. Responsibilities: Employed ResNet for image embeddings and YOLO for face detection.
Key outcomes:
Developed an image analysis script for a tree planting campaign.
Kiran
LLM