App LogoApp name
BS

Bhoomi Shakya

Open to work2 years experience

Machine Learning Engineer

Mumbai, Maharashtra, IndiaTarget Roles: Backend Engineer • Full-Stack Developer • Frontend Specialist

Machine Learning Engineer specializing in credit risk, LLMs, and agentic AI.

GitHub

Standing Rank

Rank Not Available

Developer Badges

No badges earned yet

Skills & Technologies

40 skills
javapythonhtmlcssjavascriptsqlpandasnumpyscikit-learnreactjsfastapitailwindchartjsnodejsexpressjsmachine learningnatural language processing (nlp)azure cognitive serviceslangflowlanggraphlangchainagentic aiapplication scorecardbehavioral scorecardmcpn8nrisk modelingvectorless ragmongodbmysqlvectordbazureazure app servicesazure blob storagegitgithubci/cdgithub actionsdockerAgentic AI

Work Experience

Machine Learning Engineer

Monsoon CreditTech

Nov 2025 - Present Mumbai, Maharashtra
  • Working as a Forward Deployed Machine Learning Engineer with ICICI Bank, partnering with model development and validation teams to build credit risk underwriting solutions, bureau data pipelines, model validation frameworks (Gini, KS, Lift, Decile Analysis, SHAP, OOT), and production monitoring using PSI/CSI.
  • Improved scorecard model performance from 0.40 to 0.46 Gini through feature engineering, feature selection (Sparse PCA, RFE, VIF) and Optuna-based hyperparameter tuning, resulting in improved model discrimination and underwriting effectiveness.
  • Evaluated model explainability and feature importance using SHAP, supporting business stakeholders and validation teams in interpreting model behavior and ensuring regulatory compliance.
  • Evaluated and benchmarked open-source SLMs/LLMs (Qwen 2.5, Qwen 3, and Gemma) for an Early Warning Signal (EWS) solution, enabling migration from GPT-4o Mini to open-source models and significantly reducing inference costs while maintaining comparable performance.

Associate Data Scientist

Meridian Solutions Pvt. Ltd.

Aug 2024 - Oct 2025 Gurugram, Haryana
  • Designed and deployed intelligent, data-driven systems using LLMs, agentic workflows, and custom pipelines to solve real-world problems across domains like education, finance, and analytics.
  • Built scalable AI architectures, optimizing data pipelines, and integrating models into production environments with a strong focus on performance, reliability, and automation.
  • Delivered AI-driven platforms with React.js, FastAPI, Azure, and LLMs (GPT-4, GPT-4o). Key projects include an LMS with RAG-based quiz/chatbot, and Omega Proctor AI with YOLO-based face detection and exam security features.
  • Implemented scalable CI/CD using Azure App Services and GitHub Workflows. Collaborated in agile teams to deliver production-ready solutions integrating Computer Vision, NLP, and full-stack systems.

Projects

MULTI-AGENT PROJECT TEMPLATE GENERATOR

View Project
Google ADKGeminiPythonShellLLMs
  • Developed an AI-powered system using Google’s ADK that automates project setup based on user-defined tech stacks. Implemented modular agents for tech stack planning, file system creation, validation, refactoring, and auto-generating professional README.md files.
  • Integrated tools like run_command_interactive, create_file, and exit_loop to enable a self-healing workflow (max 5 iterations), isolating all file operations in an output/ directory to ensure safe, reproducible, production-ready template generation.

EDULEARN.AI

ReactJSNode.jsMongoDBFastAPIGPT-4Azure
  • Designed and deployed an AI-enhanced LMS platform featuring GPT-4/GPT-4o-powered quiz generation, AI note creation, and a real- time chatbot, integrated with RAG pipelines and deployed via Azure Static Web Apps with automated GitHub CI/CD.
  • Built Omega Proctor AI, an exam monitoring tool with YOLO-based face detection, ID mismatch detection, sound monitoring, and tab- switch tracking, ensuring academic integrity in online environments

Leaderboard Standings

Leaderboard Position Pending

Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.

Assessment Highlights

Assessments Not Completed

Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.

AI Collaboration Score

AI Collaboration Score Pending

Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.

Role Compatibility Profile

Role Compatibility Analysis Pending

Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.

Achievements

Microsoft Azure AI-102 Certification

Earned the Microsoft Azure AI-102: Azure AI Engineer Associate certification.

LeetCode Achievement

Solved 500+ Problems based on Data Structure and Algorithms on LeetCode and 15+ contests.

About Details

Professional Bio

Machine Learning Engineer with expertise in credit risk modeling, LLMs, and agentic workflows. Experienced in deploying scalable AI solutions and optimizing data pipelines. Proven track record in improving model performance and reducing inference costs.

Bachelor of Technology (B. Tech) in AIML

GLA University (2021 - 2025)

12th Science in PCM

Dr. MPS World School ( - 2021)

10th

Police Modern Secondary School ( - 2019)

Languages: English, Hindi