Help & Supportcareers@gravityer.com
AI Engineer with 2+ years in LLM & RAG Systems, Chatbot, and Voice Bot Development
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 3 years of hands-on experience designing, building, and deploying production-grade AI systems including LLM-powered chatbots, voice bots, RAG pipelines, and agentic AI workflows. Proven expertise in Generative AI, NLP, STT/TTS fine-tuning, voice cloning, SQL analytics agents, and coding agents. Experienced with LangChain, Hugging Face Transformers, PyTorch, vector databases (Pinecone, FAISS, ChromaDB), and cloud infrastructure (AWS). Adept at architecting low-latency, scalable AI pipelines for contact center automation, enterprise knowledge retrieval, wearable health monitoring, and multilingual transcription across 5+ Indian languages.
Annai Mathammal Sheela Engineering College
Bachelor of Engineering · Computer Science
August 1, 2019 – June 30, 2023
Worktual Innovations Pvt Ltd
AI Engineer
November 1, 2023 – Present
Chennai, Tamil Nadu, India
RAG-Powered Enterprise Chatbot
November 1, 2023 – June 30, 2026
Architected a production-ready LLM + RAG system using Pinecone vector database and semantic embeddings for a 10,000+ document enterprise knowledge base. Applied advanced chunking, re-ranking, and prompt engineering, achieving 30% improvement in retrieval accuracy and near-zero hallucination rate. Shipped via REST API with real-time streaming, supporting 100+ concurrent users.
Multilingual Voice Bot for Contact Centers
November 1, 2023 – June 30, 2026
Delivered an end-to-end voice bot combining Whisper-based STT, LLM-driven intent detection, dynamic dialogue management, and neural TTS across 5+ Indian languages. Replaced legacy IVR flows, reducing average handling time by 35% and cutting operational costs by 20%.
AI Chatbot with Agentic Capabilities
November 1, 2023 – June 30, 2026
Engineered a multi-turn conversational chatbot using LangChain tool-calling (Agentic AI) to automate 3+ enterprise workflows including live data lookup, document retrieval, and appointment scheduling. Deployed across web and CRM channels, cutting manual task resolution time by 50%.
Voice Cloning & Personalized TTS System
November 1, 2023 – June 30, 2026
Designed a voice cloning pipeline using deep learning (PyTorch) to replicate speaker voices from as few as 5 minutes of audio with high fidelity. Rolled out across product personalization and branded voice assistant applications, reducing custom voice production time by 80%.
Real-Time Meeting Summarization Pipeline
November 1, 2023 – June 30, 2026
Constructed a real-time pipeline combining WebSocket-based live STT, speaker diarization, and LLM-based summarization to produce structured meeting notes and action items. Achieved end-to-end latency under 2 seconds for meetings up to 60 minutes.
Smartwatch-Based Fall Detection System
November 1, 2023 – June 30, 2026
Planned a wearable ML pipeline using accelerometer and gyroscope sensor data from smartwatches to detect falls in real time. Trained activity classification models (Random Forest, LSTM) on annotated motion datasets, achieving 92% detection accuracy. Integrated REST API alerting to notify caregivers within 3 seconds of a detected fall, improving safety outcomes for elderly users.
Natural Language to SQL Analytics Agent
November 1, 2023 – June 30, 2026
Built a NL-to-SQL agent using LangChain and GPT-4 that converts plain-English business queries into efficient SQL statements against enterprise relational databases. Integrated schema-aware prompt engineering, query validation, and error-correction loops, reducing analyst query turnaround time by 60% and enabling non-technical stakeholders to self-serve insights.
AI Coding Agent
November 1, 2023 – June 30, 2026
Created an agentic coding assistant using LangChain with code execution, retrieval, and debugging tools to automate code generation, code review, and technical documentation tasks. Integrated with developer IDEs and APIs, reducing developer task completion time by 45% and automating repetitive engineering workflows.
Fine-Tuned STT Model for Domain-Specific Transcription
November 1, 2023 – June 30, 2026
Fine-tuned OpenAI Whisper on domain-specific corpora (technical, medical, and contact center audio) using Hugging Face Transformers and custom data augmentation pipelines. Achieved a 38% reduction in Word Error Rate (WER) over the base model on in-domain test sets, improving transcription accuracy for specialized vocabulary and noisy telephony audio.
Fine-Tuned TTS Model for Branded Voice Synthesis
November 1, 2023 – June 30, 2026
Fine-tuned a TTS model (Orpheus TTS) on a natural, human-like branded voice dataset to produce natural, expressive speech aligned with product tone and persona. Applied speaker adaptation techniques and prosody control, validating results in human evaluation and deploying across IVR and voice assistant channels.
Cultural Fit Analysis
The candidate's diverse project portfolio, covering RAG systems, multilingual voice bots, voice cloning, NL-to-SQL agents, and agentic coding assistants, indicates a broad interest and capability in various AI domains. This diversity, coupled with experience in deploying solutions across multiple channels (web, CRM, telephony), suggests a versatile and adaptable individual who could integrate well into a dynamic AI engineering team. The focus on practical, impactful solutions aligns with a results-driven culture.
Soft Skills & Operational Fit
The candidate's project descriptions and professional experience highlight strong problem-solving skills, a results-oriented approach, and the ability to work on complex, end-to-end AI system development. The focus on real-world applications and measurable outcomes suggests a practical and effective operational fit. However, without direct interview data, specific soft skills like teamwork, leadership, and adaptability cannot be fully assessed.