Experiences
My journey as a software developer over 1+ years
- UPUprise Labs Private LimitedFull-Time
AI Engineer
July 2025 - Present
Key Responsibilities
- Contributing to AI-driven voice and video interview modules for Gappeo.ai, an AI hiring platform, improving candidate evaluation processes.
- Optimized voice interview workflows built on VAPI with dynamic assistant configuration, improving scalability.
- Improved video interview evaluation accuracy from 70% to 85% by implementing a QA-based pair evaluation system.
- Built an AI-driven proctoring system for technical assessments and video interviews that detects prohibited objects, multiple people, and head pose using YOLO and Azure head-pose models.
- Implemented a queue-based calling system for LiveKit-powered voice agents to manage call concurrency limits and ensure reliable call execution at scale.
- Developed and configured multilingual voice agents for diverse business use cases, optimizing language-specific prompts and speech configurations for performance and cost efficiency.
- Integrated automated testing of LiveKit-based voice agents using LiveKit Agent Simulations, designing positive and negative test scenarios to validate agent behaviour, response accuracy, and edge-case handling.
- Built and scaled voice AI agents across multiple business use cases, handling an average of 10,000+ conversation minutes per day while maintaining P95 latency of 2 seconds.
- Improved speech recognition and synthesis quality by implementing context-aware STT prompting and TTS pronunciation dictionaries, enhancing recognition of domain-specific terminology and ensuring accurate pronunciation of names, technical terms, and custom vocabulary.
- Developed a visual, workflow-based automation module similar to n8n and Zapier, enabling recruiters to configure hiring stages using drag-and-drop nodes, define conditional rules, and automate recruitment workflows based on predefined conditions.
- Improved video call performance with noise cancellation, tool calling, and voice activity detection — reducing latency by 40% and improving transcript quality by 30% using AssemblyAI.
- Built AI agents for candidate smart search and job creation using prompt-driven workflows, achieving response times under 2 seconds.
- Implemented an AI-assisted interview co-pilot that observes live conversations and recommends the next questions to recruiters in real time (<2s latency), integrated with interview scheduling and Google Calendar sync.
- Developed a low-latency WebSocket-based real-time messaging system for recruiters and candidates — messages, file sharing, live notifications, and online/offline presence — unifying external and internal chat workflows.
- Built a dynamic AI voice interview agent that adapts questions based on candidate responses and recruiter-defined rules, with tool calling for call termination, email notifications, and custom recruiter actions.
Technology Used
PythonFastAPIAI AgentsLiveKitVAPIWebRTC
- LELeadBird.ioInternship
Software Engineering Intern
March 2025 - July 2025
Key Responsibilities
- Developed Python automation solutions integrating external APIs, Airtable, and Smartlead for streamlined workflows.
- Built robust full-stack apps with Lovable.ai integrated with FastAPI backend APIs.
- Designed and implemented Zapier automation workflows connecting third-party apps including Smartlead, and optimized existing implementations for performance and scalability.
- Implemented asynchronous programming methods to improve API efficiency.
- Actively supported Customer Success Managers (CSMs) by resolving technical queries and providing reliable solutions.
Technology Used
PythonFastAPIZapierAirtableLovable.ai