08/06/2026 updated
SG
100 % available
Data Analyst & AI/ML Engineer | Python, SQL, LLM & RAG Systems
Ghaziabad, India
Worldwide
Master of Science in Data Management for Corporate FinanceGCPApplication Programming Interfaces (APIs)Business Intelligence DevelopmentGoogle BigQueryExtract Transform Load (ETL)Data ManagementGitHubPython (Programming Language)PostgreSQLMachine LearningAzure Machine LearningSQL DatabasesStakeholder ManagementWorkflowsBusiness Intelligence Development Studio
AI & Machine Learning Systems
Expertise in building and deploying AI pipelines, LLM orchestrations, RAG systems, and agentic workflows using tools such as LangChain, Ollama, FAISS vector databases, and Model Context Protocol (MCP) servers. Experience shipping production AI feature modules and benchmarking machine learning models for financial applications.
Data Engineering & Pipeline Orchestration
Strong background in designing and maintaining end-to-end data pipelines using Databricks, PySpark, dbt Framework, and ETL processes. Proven track record of optimizing batch model inference processing times and orchestrating production machine learning pipelines at scale.
Programming & Backend Development
Proficiency in Python (FastAPI, Pydantic, Pandas, NumPy, Scikit-learn, PySpark, Pytest), SQL, PostgreSQL, and REST APIs. Experience developing backend automation, API handlers, and asynchronous systems using AWS Step Functions, Lambda, and S3.
Database & Query Optimization
Development and optimization of complex backend data layers using PostgreSQL, including high-performance SQL queries, structural schemas, and stored procedures for lending and campaign analytics.
Cloud & Infrastructure Tools
Hands-on experience with AWS services (Lambda, S3, Step Functions, API Gateway), Docker, Databricks, BigQuery, and Git/GitHub for infrastructure monitoring and deployment of high-volume production systems.
Statistical Analysis & Evaluation Frameworks
Design and implementation of automated statistical testing frameworks (Chi-square based) to monitor feature store stability and detect upstream data quality anomalies before they impact downstream ML models.
Data Visualization & Reporting
Utilization of Power BI, Tableau, and Mixpanel for financial data visualization and reporting. Experience building internal functional data frameworks and end-to-end user workflows using Python and Figma to reduce manual reporting loops.
Document Processing & Intelligent Document Processing (IDP)
Building document processing applications using FastAPI, LangChain, and Ollama to parse complex financial PDFs. Structuring raw documents using Docling into clean Markdown representations to maximize extraction quality.
Expertise in building and deploying AI pipelines, LLM orchestrations, RAG systems, and agentic workflows using tools such as LangChain, Ollama, FAISS vector databases, and Model Context Protocol (MCP) servers. Experience shipping production AI feature modules and benchmarking machine learning models for financial applications.
Data Engineering & Pipeline Orchestration
Strong background in designing and maintaining end-to-end data pipelines using Databricks, PySpark, dbt Framework, and ETL processes. Proven track record of optimizing batch model inference processing times and orchestrating production machine learning pipelines at scale.
Programming & Backend Development
Proficiency in Python (FastAPI, Pydantic, Pandas, NumPy, Scikit-learn, PySpark, Pytest), SQL, PostgreSQL, and REST APIs. Experience developing backend automation, API handlers, and asynchronous systems using AWS Step Functions, Lambda, and S3.
Database & Query Optimization
Development and optimization of complex backend data layers using PostgreSQL, including high-performance SQL queries, structural schemas, and stored procedures for lending and campaign analytics.
Cloud & Infrastructure Tools
Hands-on experience with AWS services (Lambda, S3, Step Functions, API Gateway), Docker, Databricks, BigQuery, and Git/GitHub for infrastructure monitoring and deployment of high-volume production systems.
Statistical Analysis & Evaluation Frameworks
Design and implementation of automated statistical testing frameworks (Chi-square based) to monitor feature store stability and detect upstream data quality anomalies before they impact downstream ML models.
Data Visualization & Reporting
Utilization of Power BI, Tableau, and Mixpanel for financial data visualization and reporting. Experience building internal functional data frameworks and end-to-end user workflows using Python and Figma to reduce manual reporting loops.
Document Processing & Intelligent Document Processing (IDP)
Building document processing applications using FastAPI, LangChain, and Ollama to parse complex financial PDFs. Structuring raw documents using Docling into clean Markdown representations to maximize extraction quality.
Experience
Professional experience0 - 5 years
Freelancer since2023
Languages
EnglishFluentHindiNative speaker
Project history
Built a document processing application using FastAPI, LangChain, and Ollama to parse complex financial PDFs. Developed local RAG pipelines utilizing DeepSeek-R1 embeddings and FAISS vector databases for semantic retrieval. Structured raw, unformatted documents using Docling into clean Markdown representations to maximize extraction quality and decrease model hallucinations.
Engineered a prototype production AI Data Analyst using Model Context Protocol (MCP) servers to link Claude LLM orchestrations with BigQuery data warehouses. Shipped and optimized 3 production AI feature modules. Built internal functional data frameworks and end-to-end user workflows using Python and Figma, reducing manual reporting loops by 60%. Diagnosed a multi-country asynchronous billing logic flaw and wrote a patch using Python and Twilio APIs.
Designed and built an end-to-end automated statistical testing framework (Chi-square based) to monitor feature store stability. Developed and optimized complex backend data layers using PostgreSQL. Maintained and calibrated production machine learning pipelines and Databricks Orchestrator routines, refactoring PySpark data handling to slash batch model inference processing times by 70%. Supported deployment, integration, and continuous infrastructure monitoring of multiple high-volume production ML models.