S

Shubhankit Sirvaiya

AI Researcher

Gurugram, Haryana, India5 yrs 10 mos experience
Most Likely To SwitchAI Enabled

Key Highlights

  • Designed AI solutions for diverse industries.
  • Achieved 98% accuracy in automated invoice processing.
  • Developed scalable MLOps frameworks for vehicle servicing.
Stackforce AI infers this person is a Data Science and MLOps expert in Fintech and SaaS industries.

Contact

Skills

Core Skills

MlopsGenerative AiNatural Language Processing (nlp)Data ScienceMachine Learning

Other Skills

AlgorithmsAnalyticsArtificial Intelligence (AI)Azure APIMAzure Copilot StudioAzure Data FactoryAzure DatabrickAzure FunctionsAzure OpenAIBusiness Intelligence (BI)CI/CD pipelinesContinuous Delivery (CD)Continuous Integration (CI)Convolutional Neural Networks (CNN)Dashboard

About

As a Machine Learning Engineer at World Wide Technology, I design and implement AI solutions that empower businesses to optimize their performance and achieve their goals. I have over 5 years of experience in Data Science, Business Intelligence, and Consulting, where I have applied my skills and expertise in various domains such as telecom, automobile, and energy. One of my recent projects involved deploying a flexible and hybrid end-to-end MLOPs framework using Azure Databrick, Azure Data Factory, Azure Functions, and CI/CD pipelines to predict the labor hours required for vehicle servicing. This solution enabled the client to reduce operational costs, improve customer satisfaction, and increase efficiency. I am passionate about contributing to the AI revolution that is transforming the way businesses operate, and I am always eager to learn new technologies and techniques that can enhance my capabilities and impact. I have completed multiple certifications and courses in Data Science, such as the Machine Learning Specialization by deeplearning.ai, the SQL fundamentals by SoloLearn, and the Excel for business essential by Coursera. I have also achieved the Kaggle notebook expert status and the top data mining voice 2024 recognition on LinkedIn. I enjoy sharing my work and insights on various platforms, such as Kaggle, GitHub, Medium, and Twitter, where I connect with a vibrant community of data enthusiasts. I am enthusiastic about new projects and opportunities that allow me to leverage my skills and expertise and work alongside teams that are making an impact in the industry. Please feel free to reach out to me for any queries or collaboration opportunities.

Experience

5 yrs 10 mos
Total Experience
1 yr 11 mos
Average Tenure
2 yrs 8 mos
Current Experience

World wide technology

AI Engineer

Oct 2023Present · 2 yrs 8 mos · Remote

  • Designed an intelligent retrieval agent to automate extraction and
  • summarization of SEC Edgar filings and financial data to support internal research, compliance, and decision-making. Developed an end-to-end solution combining Azure Copilot Studio, Power Automate, LangChain, and Azure OpenAI , including custom connectors to
  • fetch submissions and facts, dynamic routing logic, and summarization pipelines. Faced challenges managing large JSON payloads exceeding token limits ; resolved by implementing dynamic filtering and chunking strategies , reducing payload size by 60% , and optimizing processing time. Enabled teams to retrieve and analyze complex SEC data in under 30
  • seconds , automating 70%+ of manual document
  • Invoice Auditor GenAI-OCR Model (02/2024 - 12/2024)
  • Designed and deployed an automated invoice processing pipeline using
  • Azure OpenAI, Azure APIM, and prompt engineering, achieving 98%
  • accuracy in data extraction for structured text conversion.
  • Integrated OCR methods and OpenCV for PDF-to-text conversion, reducing manual invoice processing time by 40% across 7+ clients . Deployed the solution via FastAPI , Docker , and Kubernetes , with CI/CD
  • automation using Jenkins , cutting deployment overhead by 30% . Delivered scalable, production-ready systems that achieved up to 25% cost savings and improved processing efficiency for customer workflows.
  • Chatfusion- Contextual Chatbot Implementation
  • (04/2024 - 06/2024)
  • Developed ChatFusion, a multi-turn contextual chatbot using transformer- based LLMs and LangChain, enhancing conversational accuracy by 20%. Engineered advanced NLP pipelines with SpaCy , NLTK , and embedding- based retrieval, reducing query resolution time by 15% . Deployed the chatbot with FastAPI and Docker , ensuring 99.5% uptime and
  • handling 10,000+ daily queries with robust scaling via Kubernetes.
  • Improved chatbot performance through hyperparameter tuning and API monitoring, leading to higher adoption rates and user satisfaction scores.
LangchainLarge Language Models (LLM)Generative AIFastAPIMLOpsContinuous Integration (CI)+2

Accenture ai

Data Scientist

Oct 2021Oct 2023 · 2 yrs · Gurugram, Haryana, India

  • Cashflow Optimization Tool
  • Tech Stack - Python, Pandas, NumPy, Pulp , Django , SQL
  • Designed the tool to analyze data, retrieve information using SQL queries, and model cash flow scenarios. By employing advanced mathematical models, the tool has the potential to optimize cash flow by up to 35%, enabling businesses to make data-driven decisions and maximize their financial performance.
  • The tool's intuitive interface allows businesses to input their financial data and receive actionable insights for cash flow optimization.
  • Labour Hour Prediction For Vehicle Servicing
  • Tech Stack - MLOps, Azure Databrick, Azure Data Factory,Azure Functions. CI/CD pipelines,Python, SQL, ML.
  • Deployed a flexible and hybrid end-to-end MLOP s (CI/CD) Big Data modelling architecture using Azure cloud platform, Spark and DataBricks .
  • Created a serverless scalable microservice using Azure functions and reduce the time latency by 5x. Consulted major Japanese Farm equipment manufacturer with machine learning and AI-driven insights and solutions.
  • Developed KPIs and a chained gradient-based boosting ensemble model to predict the labor hours required to complete the job.
  • Zero Based Saving analysis tool
  • Tech Stack- Python, SQL, ML, Django, HTML, PowerBI, Excel.
  • Developed a Zero Based Saving analysis tool, which was built on Django Framework. Worked on analysis using Python libraries and looked out for possibilities of including Machine Learning techniques across different Initiatives in the tool.
  • Implemented Data Modelling , Clustering and Optimization techniques. Gained the understanding of Energy domain , also used the basics of Frontend tools HTML, CSS and JavaScript.
Data StructuresNeural NetworksData VisualizationSQLPython (Programming Language)Artificial Intelligence (AI)+5

The sparks foundation

Data Science and Analytics Internship

Apr 2021Apr 2021 · 0 mo

  • Applied Supervised and Unsupervised Algorithm for Multiclass Classification Problem. Used and Visualized Decision Tree Algorithm in Class Prediction. Made Interactive Dashboards using PowerBI for Data Visualization and EDA.
Natural Language Processing (NLP)

Quantium

Intern

Mar 2021Apr 2021 · 1 mo

  • Completed Data Analytics Virtual Experience Program
  • Completed Following Task-
  • Data Preparation and Customer analytics Using Pandas Numpy and
  • Matplotlib Libraries.
  • Experimentation and Uplift testing Analytics Using Pearson Correlation followed by commercial application report showing insights gained.

Ge

Intern

Mar 2021Mar 2021 · 0 mo · Virtual

  • Virtual Internship at General Electric Under GE Digital Technology Data Analytics Program.
  • This program included combining and visualizing data to drive meaningful insights using enterprise-grade technology.
  • Here I learned Deep Insights about Data Engineering ,Data Visualization using Excel and Dataiku.

Cubastion consulting

Associate Data Scientist

Jul 2020Sep 2021 · 1 yr 2 mos · Gurugram, Haryana, India

  • -Consulted major Indian Telecom organizations. Developed and integrated the Customer Segmentation model in the CRM using demographic and behavioral stats. -Improved the model performance Significantly by adding polynomial features and a normalization layer. Channelized the BI pipeline for KPI metrics for different stakeholders. -Extensively used SQL for data acquisition & wrangling, Python for data processing and KPI generation, and Power BI for visualization.
Artificial Intelligence (AI)Natural Language Processing (NLP)Data Science

Jsw

Summer Intern

May 2019Jul 2019 · 2 mos · Bellary, Karnataka

  • Completed a project on steel making.
  • Involving Kanbara Reactor Process of desulphurization.
  • Used Excel for Analysis of time taken in one complete cycle and possible ways to reduce it.

Rashtriya ispat nigam limited (rinl)

Winter Intern

Dec 2018Jan 2019 · 1 mo · Vishakhapatnam, Andhra Pradesh, India

  • Winter Internship RINL,Vishakapatnam
  • Here i worked on the project Based on the working Principles of Blast Furnace Department of Steel Making.

Education

Malaviya National Institute of Technology Jaipur

Bachelor of Technology - BTech — metallurgy and material science

Jan 2016Jan 2020

Stackforce found 100+ more professionals with Mlops & Generative Ai

Explore similar profiles based on matching skills and experience