Help & Supportcareers@gravityer.com
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 1+ years in ML & GenAI Systems
AI/ML Engineer with 1+ years of experience in building and deploying production-grade ML and GenAI systems across logistics, document processing, and data extraction. Skilled in designing RAG pipelines, LLM-powered applications, and scalable cloud architectures. Hands-on experience across the ML lifecycle, including data processing, model development, deployment, monitoring, and scheduled retraining. Proven ability to deliver reliable high-impact systems that reduce manual effort and improve efficiency. Interested in extending these systems towards agentic and autonomous AI workflows for real-world production use cases.
LDRP Institute of Technology and Research, Gandhinagar, Gujarat
Bachelor of Engineering · Computer Engineering
October 1, 2021 – April 1, 2025
Karnavati English Medium School, Ahmedabad, Gujarat
Higher Secondary School (XII)
April 1, 2020 – July 1, 2021
Jay Somnath English Medium School, Ahmedabad, Gujarat
Secondary School (X)
April 1, 2018 – May 1, 2019
Tatvasoft
Jr. AI/ML Engineer
January 1, 2025 – Present
India
Delivery Intelligence Chatbot
June 24, 2026 – Present
Built a RAG-based logistics chatbot enabling operations teams to query order, invoice, and trip data in natural language, reducing manual reporting effort. Designed a hybrid retrieval system combining vector search (Weaviate) with structured database queries, reducing response latency by ~ 50% through parallel retrieval for multi-source queries. Implemented context-aware conversation memory, minimizing redundant database calls and improving multi-turn query handling. Developed a document ingestion pipeline (Azure OCR + embeddings) to enable semantic search over unstructured shipping documents. Improved system reliability and cost efficiency by implementing async processing (queues), LLM request routing, and rate limiting.
Document Ingestion & Multi-Agent Extraction System
June 24, 2026 – Present
Built an LLM-powered document processing system to extract structured data from financial documents (PDF, Excel, CSV), reducing manual data entry by ~ 90%. Designed a modular multi-stage (multi-agent) pipeline where separate components handle OCR, parsing, validation, and error correction to improve extraction accuracy (~95%) across diverse document formats. Enabled ingestion of heterogeneous document formats (scanned PDFs, native PDFs, spreadsheets), handling both structured and semi-structured layouts. Implemented LLM-based validation + rule-based consistency checks to ensure data correctness across diverse document schemas. Reduced pipeline costs by ~ 60% using selective retries, feedback loops, and failure-aware processing.
Horse Race Prediction
June 24, 2026 – Present
Developed a machine learning prediction system to forecast horse race outcomes using historical race data and advanced feature engineering (form trends, win ratios). Trained and evaluated multiple models (Logistic Regression, XGBoost, LightGBM, Random Forest), selecting the best-performing model based on validation metrics. Built automated data ingestion and preprocessing pipelines using AWS Step Functions and AWS Lambda for continuous data updates. Deployed the model using AWS SageMaker pipelines, enabling scheduled retraining, model versioning, and performance monitoring. Improved prediction reliability by incorporating feature selection, statistical analysis, and model comparison techniques.
Conversational Inventory Intelligence System
June 24, 2026 – Present
Developed an AI-powered conversational interface that enables users to query and manage enterprise inventory data using natural language, eliminating the need for manual SQL queries. Built scalable backend services using Python and FastAPI, integrated with PostgreSQL for secure and efficient inventory operations. Leveraged LangGraph, LangChain, and OpenAI LLMs to implement intent classification and safe natural language-to-SQL translation. Ensured enterprise-grade reliability through role-based access control (RBAC), multi-layer validation, human-in-the-loop approvals, and comprehensive audit logging. Containerized and deployed the system using Docker, enhancing scalability and enabling seamless integration with ERP and supply chain platforms.
Cultural Fit Analysis
The candidate's projects demonstrate a diverse application of AI/ML across different domains (logistics, document processing, financial data extraction, inventory management, predictive analytics). This breadth of experience, coupled with an interest in extending systems towards agentic AI, suggests adaptability and a proactive learning attitude. The focus on building production-grade systems and improving efficiency aligns well with a results-oriented culture. However, the candidate's experience level (Jr. AI/ML Engineer, still pursuing a Bachelor's degree) might indicate a need for mentorship and structured guidance in a senior role, which could be a factor in cultural integration for a highly autonomous team.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a strong problem-solving aptitude and a focus on delivering tangible business value (e.g., reducing manual effort, improving accuracy, reducing costs). The emphasis on reliability, scalability, and cost efficiency in project implementations suggests a practical and operationally aware mindset. The objective statement also highlights an interest in agentic and autonomous AI workflows, aligning with advanced operational trends.