Sarthak sharma

AI Researcher

Ghaziabad, Uttar Pradesh, India2 yrs 5 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in building production-ready AI systems.
Proficient in Generative AI and Large Language Models.
Skilled in automating workflows and enhancing decision-making.

Stackforce AI infers this person is a SaaS and Fintech specialist with expertise in AI-driven automation and data analysis.

Contact

Skills

Core Skills

Machine LearningGenerative AiNatural Language Processing

Other Skills

Knowledge GraphsSemantic SearchML DashboardSQL AgentMCP IntegrationPythonSQLLangGraphAutomationData AnalysisLLM AgentsData VisualizationLead ScoringFinance BotDocument Chatbot

About

Building production-ready AI and machine learning systems with a strong focus on Generative AI, Large Language Models (LLMs), Model Context Protocol (MCP), and NLP. Experienced in designing and deploying AI Agents that automate workflows, generate insights, and accelerate decision-making across data-driven processes. Proficient in creating intelligent automation pipelines using LLMs, knowledge graphs, retrieval-augmented generation (RAG), and context-aware architectures to unlock semantic search, task automation, and predictive intelligence. Skilled in enhancing operational efficiency by integrating Generative AI-based agents into analytics workflows and self-service systems that generate code, insights, and scalable solutions. Technical expertise includes Python, PyTorch, TensorFlow, LangChain, LangGraph, LlamaIndex, OpenAI, Claude, and vector databases — combined with experience in scalable model deployment, MCP-enabled pipelines, and cloud-based AI workflows. Academic foundation in Data Science and Statistics with research experience in NLP. Driven by building ethical, scalable, context-aware AI systems that automate complex workflows and empower smarter business outcomes.

Experience

Inncretech

AI Software Engineer

Feb 2025 – Present · 1 yr 2 mos · Princeton, New Jersey, United States · Remote

Designed and implemented a hybrid Knowledge Graph + Semantic Search system over PDFs and tabular data, using query planning and parallel retrieval, improving answer accuracy by 35% and reducing response latency by 30% for complex user queries.
Developed an interactive ML Dashboard enabling users to upload or connect datasets and perform machine learning analysis.
Added automated evaluation metrics for various ML techniques, leading to a 40% reduction in model testing time and 30% increase in user engagement on the platform.
Built a dynamic SQL Agent using LangGraph, capable of handling schema-driven natural language queries. Enabled features such as query generation, optimization, and debugging, resulting in a 45% improvement in query resolution speed and 25% reduction in manual SQL writing errors.
Developed intelligent automation agents for Jira, Salesforce, HubSpot, and pipelines with MCP integration and Human-in-the-Loop approval for write operations; improved automation, data accuracy, and reduced manual effort by 40%.
Built a Python Agent for data-related queries that generates Python code, tables, analysis, and Plotly charts; enabled self-service analytics and reduced manual analysis time by 50%.

Knowledge GraphsSemantic SearchML DashboardSQL AgentMCP IntegrationPython+2

Upadpro software & services pvt. ltd.

AI/ML Engineer

Nov 2023 – Feb 2025 · 1 yr 3 mos · Noida, Uttar Pradesh, India · On-site

🖥️ Developed a savedsearch bot for the savedsearch module using custom agents. Integrated LLM agents into the bot to enhance data visualization capabilities by generating charts and graphs from pandas dataframes. This integration enables more comprehensive data interpretation and facilitates intuitive communication with the bot.
📈 Developed a lead scoring and Opportunity Scoring feature in the Liono360 product, resulting in a 20% increase in product sales. This feature enables sales representatives to easily identify and prioritize high-value leads, enhancing sales efficiency and effectiveness.
💹 Developed a cutting-edge finance bot using custom agents and specialized tools for generating, analyzing, and forecasting financial statements, offering deep financial insights. The bot integrates custom software agents with advanced financial modules, enabling precise data handling and rich visualizations. This synergy allows for robust financial forecasting and interactive communication, making financial analysis more thorough and user-friendly.
📄 Created a Document chatbot capable of interacting with various types of data files, including URLs and PDFs. Implemented advanced RAG Pipeline "Sentence windows retrieval" using llama-index for efficient data retrieval and processing.
🤖 Developed a comprehensive ML model replicating AKKIO’s capabilities for the saved search module, designed to work seamlessly across diverse datasets. This project incorporates various machine learning algorithms for prediction and utilizes deep learning techniques for forecasting. The model automates data cleaning and preprocessing, ensuring data quality and readiness for analysis.

LLM AgentsData VisualizationLead ScoringFinance BotDocument ChatbotMachine Learning+1

Extend future

Machine Learning Engineer

Apr 2023 – Jul 2023 · 3 mos · Pune District, Maharashtra, India · Remote

Developed and implemented OpenAI embedding techniques for creating embeddings and stored them in the Milvus vector database.
Created scripts for managing the Milvus vector database and optimized its performance.
Built APIs using the Sanic framework for seamless integration and access to embedded data, while utilizing MongoDB and PyMongo for data uploads, modifications, and content generation using Openai.

OpenAI EmbeddingMilvus Vector DatabaseAPIsMongoDBMachine Learning

Resoluteai.in

Machine Learning Engineer

Jul 2021 – Oct 2021 · 3 mos · India

Utilized Python and implemented the CNN deep learning technique to create a proof of concepts for Suspecioius Object detection models.
Added new features to a logistics industry-based project, achieving a 90% accuracy rate in model performance.
Developed a web application for data visualization and predictive modeling for proof of concepts.

PythonCNNData VisualizationMachine Learning