Help & Supportcareers@gravityer.com
AI Engineer with 1+ years in Machine Learning, NLP & Computer Vision
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Junior AI Engineer with hands-on production experience in Machine Learning, NLP, Computer vision and Reinforcement Learning. Proven track record building end-to-end ML pipelines, deploying real-time inference systems, and integrating LLM-based solutions. Proficient in Python, CVAT, TensorFlow, Hugging Face Transformers, LangChain, Flask/FastAPI. Published research paper in Insider threat detection that uses LSTM Autoencoder along with Gaussian Mixture model to range threats from 0 - 100. A Technician, designer and operator of bots at Raven Robotics.
GH Raisoni University
B.Tech · Artificial Intelligence
August 1, 2022 – June 30, 2026
Karmik Nexus IT Solutions
Computer Vision Engineer
December 1, 2025 – June 30, 2026
Nagpur, Maharashtra, India
The Dexter
AI intern
March 1, 2025 – May 31, 2025
Delhi, Delhi, India
Orinsons Technologies Private Limited
Machine Learning intern
December 1, 2024 – January 31, 2025
Raipur, Chhattisgarh, India
Smart Road Management System
June 1, 2026 – June 30, 2026
Developed Smart Road Intelligence System an AI-powered platform for automated road inspection and pothole detection using YOLOv8 on video feeds, with geospatial risk assessment via OSMnx and Folium. Implemented real-time pothole detection, repair cost estimation, and contractor alert system using Twilio SMS; built an Admin Command Center for city planners with global analytics and data management.
Chakrashield - AI-Powered Insider Threat Detection System
June 1, 2026 – June 30, 2026
Architected an end-to-end insider threat detection system combining LSTM Autoencoder (sequential anomaly detection) with Gaussian Mixture Model risk scoring, producing a real-time threat index (0-100) with sub-50ms inference latency. Implemented unsupervised anomaly detection on sequential user-activity data; system correctly flagged 100% of threat cases in the held-out test set; published findings in peer-reviewed research. Designed scalable data processing pipeline for ingestion, embedding, and retrieval of activity logs - combining unsupervised ML with domain heuristics for robust, explainable risk scoring. Built an interactive Streamlit + Plotly dashboard with user-behaviour heatmaps, anomaly timelines, peer-comparison charts, and graph-based threat escalation flows, enabling security teams to perform drill-down analysis.
GreenGold - RL-Driven Datacenter Intelligence Platform
June 1, 2026 – June 30, 2026
Designed a Q-Learning / DQN Reinforcement Learning agent (MDP: 48 states, 4 actions, multi-objective reward) to autonomously optimise GPU workload distribution across 5 geographically distributed regions without pre-programmed rules. Agent reduced thermal anomalies from 35+ overheating chips to 0 and stabilised average GPU temperature from 82°C to 53°C by episode 1,300; average reward converged to a stable +0.63. Achieved carbon emission reduction by migrating workloads from high-carbon regions (650 gCO2/kWh) to renewable-energy regions (11 gCO2/kWh), demonstrating multi-objective optimisation under real-world constraints. Built a Flask REST API (5 endpoints) with a live Chart.js dashboard; designed an offline analytics pipeline covering data generation, DQN training, migration planning, and automated financial audit reporting; deployed via Gunicorn.
Attention Mechanisms, Encoder-Decoder Architecture, Prompt Design in Vertex AI
Google Cloud
June 1, 2026 – Present
Tantrafeista (IIIT)
Hackathon
June 1, 2026 – Present
Meta Hackathon (Scaler School of Technology)
Hackathon
June 1, 2026 – Present
Autonomous Vehicle Workshop
RoboLearnIndia LLP
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's diverse academic projects (insider threat, smart roads, datacenter optimization) and professional experience (computer vision, generative AI) demonstrate a broad interest in applying AI across different sectors. Their participation in hackathons and leadership roles indicates a proactive and engaged approach, which aligns well with a dynamic, innovation-driven culture. The focus on explainable AI and multi-objective optimization suggests a thoughtful approach to ethical and practical considerations in AI development.
Soft Skills & Operational Fit
The candidate's project descriptions highlight strong problem-solving skills, an ability to work on complex, multi-objective problems, and a focus on delivering practical, deployable solutions. Their involvement in hackathons and a public speaking club suggests good collaboration and communication potential. The experience in designing interactive dashboards indicates an understanding of user needs and operational deployment.