AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer with less than a year in LLMs, Voice AI, and RAG Pipelines
Information Technology student with 6 months of hands-on AI/ML internship experience at F22 Labs, worked on Large Language Models (LLMs), Voice AI Agents, RAG (Retrieval-Augmented Generation) pipelines, chunking strategies, workflow automation, document extraction, structured data extraction, model evaluation, prompt engineering, and AI-powered automation solutions. Experienced in testing and analyzing open-source AI models, building extraction and automation pipelines, integrating REST APIs, and debugging end-to-end AI workflows. Passionate about leveraging technology to solve real-world problems and seeking opportunities to expand my technical expertise, learn from industry professionals, gain exposure to diverse engineering challenges, and build a strong foundation for a successful career while contributing effectively to organizational growth and innovation.
Anand Institute of Higher Technology
B.Tech · Information Technology
August 1, 2022 – June 30, 2026
Model Matriculation School
HSC (Class XII)
N/A – May 31, 2020
Model Matriculation School
SSLC (Class X)
N/A – May 31, 2022
F22 Labs
AI/ML Engineering Intern
July 1, 2025 – June 1, 2026
India
AI-Powered Resume Screening System
January 1, 2025 – June 1, 2026
Developed an end-to-end AI-driven resume screening system, automating PDF extraction, candidate profiling, Neo4j graph storage, semantic search, and hybrid graph + embedding-based retrieval to identify and rank the most relevant candidates based on job requirements and resume content. Implemented RAG pipelines using Jina Embeddings and Neo4j to enable intelligent candidate matching, contextual retrieval, structured querying, and natural language explanations for candidate recommendations. Evaluated and benchmarked GPT-4o and Qwen3-32B on 100+ resumes, comparing extraction quality, candidate-matching accuracy, processing speed, and cost efficiency to support optimal model selection. Demonstrated cost-effective and scalable resume screening workflows through model benchmarking and performance analysis, enabling efficient identification of the best-fit candidates for hiring decisions.
AI Voice Interview Agent - Harness Engineering
January 1, 2025 – June 1, 2026
Built a voice interview agent using LiveKit (real-time streaming), Deepgram nova-3 (STT), Groq llama-3.3-70b (LLM), and Cartesia sonic-3.5 (TTS). Implemented a LangGraph harness to route off-topic, reschedule, and email/JD requests without invoking the LLM — reducing latency and token cost. Measured significant latency improvements: harness-routed flows (802-950 ms avg) vs. LLM-only flow (1340 ms avg) — up to 40% faster for common request types. Used Silero VAD for real-time voice activity detection to manage conversation turn-taking accurately.
Remote Health Monitoring for Rural Patients [Final Year Project]
January 1, 2025 – June 1, 2026
Designed and built an IoT-based health monitoring system collecting vital parameters (heart rate, temperature, BP, SpO2, glucose) using wearable sensors and an ESP32 microcontroller. Developed a Python Flask backend to receive sensor data over Wi-Fi, store records in SQLite, and classify risk levels (Normal / Medium / High) against medical thresholds. Integrated Twilio API to automatically trigger emergency phone calls and SMS alerts to doctors/caregivers for critical threshold breaches. Built a web dashboard for remote real-time and historical patient health data monitoring with video consultation support.
Multilingual Voice Agent
January 1, 2025 – June 1, 2026
Benchmarked and compared multiple STT/TTS stacks for multilingual voice AI applications, evaluating latency, accuracy, and language coverage across Tamil, Telugu, Malayalam, Kannada, Hindi, and English. Conducted end-to-end performance testing of Deepgram, Sarvam Saaras v3, Cartesia, and Bulbul v3, identifying Sarvam-based pipelines as the most efficient solution with lower overall latency (~1.42 seconds) and comprehensive Indian language support. Analyzed benchmark results and provided data-driven recommendations for production-grade multilingual voice AI deployments based on performance, accuracy, and user experience metrics.
Pre-Chunking vs Post-Chunking Strategies in RAG Pipelines
F22 Labs
January 1, 2026 – Present
Self-Consistency Prompting in Large Language Models
F22 Labs
January 1, 2026 – Present
Cultural Fit Analysis
The candidate's project diversity, ranging from resume screening to voice agents and IoT health monitoring, indicates a broad interest in applying AI to various domains. Their focus on optimizing for cost and latency, along with benchmarking different models, aligns with a culture that values efficiency and data-driven decision-making. The publication of technical blogs suggests a willingness to share knowledge and contribute to the community, which is a positive cultural indicator. The candidate's academic background combined with a significant internship in AI/ML engineering shows a clear alignment with an AI Engineer role.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through debugging and optimizing complex multi-component AI systems. Their data-driven approach to benchmarking and model selection indicates a methodical and analytical mindset. The ability to self-teach new technologies quickly (LiveKit, RAG, Neo4j, LangGraph) suggests adaptability and a proactive learning attitude, which are valuable for operational fit in a fast-paced AI engineering environment. The candidate's project descriptions highlight an understanding of real-world trade-offs in latency, cost, and reliability, crucial for production-grade AI systems.