--- title: "Shubham Goyal - Manager / AI architect ( Data & Infra) in Noida | Tal" description: "Meet Shubham Goyal, Enterprise GenAI & Big Data Architect | LLM Systems & RAG Architect | Data Engineer| Building Scalable GenAI Systems (2.5M+ tokens/day, <​50ms latency | Capgemini & Cisco 2026 AI excellence award winner based in Noida. AI stack includes antigravity and opencode. Explore public w..." image: "https://tal.club/meet/shubham-goyal/share-image" --- # Shubham Goyal on Tal ## Headline Enterprise GenAI & Big Data Architect \| LLM Systems & RAG Architect \| Data Engineer\| Building Scalable GenAI Systems \(2\.5M\+ tokens/day, <​50ms latency \| Capgemini & Cisco 2026 AI excellence award winner ## About Enterprise GenAI & Big Data Architect with 6\.5\+ years of experience specializing in the design of production\-grade enterprise GenAI platforms and high\-performance distributed data foundations \. Expert in bridging high\-throughput multi\-agent orchestration with robust data engineering architectures to deliver sub\-second latency AI solutions\. Tech Stack & Domain Expertise:GenAI Architecture & LLMOps: Azure OpenAI, LangGraph, LangChain, LlamaIndex, Agentic RAG, Semantic Routing & Caching, Vector Databases \(ChromaDB, FAISS\), Guardrails & LLM Security \.Data Platforms & Distributed Systems: PySpark \(Apache Spark\), Databricks \(Medallion Architecture\), Snowflake, Delta Lake, Distributed Processing & Shuffle Optimization\. Cloud & Infrastructure: AWS \(EKS, EC2\), Azure \(ADLS Gen2\), Apache Airflow, Docker, CI/CD pipelines, MLflow Career Highlights & Impact:Enterprise GenAI Platform: Architected a production\-grade multi\-agent orchestration platform using LangGraph & Azure OpenAI serving 1,000\+ users, driving $500K\+ in annual savings\. Latency Reduction: Engineered an NLP\-driven semantic routing layer and Redis\-backed semantic caching, dropping query latency from 15s to <50ms \. High\-Scale Data Pipelines: Spearheaded a multi\-cloud Python & SQL ETL pipeline processing 112M\+ clinical records across AWS and Azure ## Location Noida ## AI stack antigravity opencode ## Work experience Manager / AI architect \( Data & Infra\) at Capgemini, Aug 2025 \- Present · 1 yr 1 mo, Noida, Uttar Pradesh, India, 1\. End\-to\-End Generative AI Pipeline Architecture \- Architected and developed an end\-to\-end Generative AI data pipeline to automate the creation of a Product Tech Specs Database from unstructured web data \(HTML/PDFs\)\. This system autonomously processed thousands of network hardware SKUs to build a centralized, query\-ready knowledge base, serving as the foundation for a strategic AI agent initiative for sales and engineering teams\. 2\. Advanced Prompt Engineering & Structured Data Extraction \- Engineered a sophisticated data extraction module using LangChain, Pydantic, and a Cisco\-hosted Azure OpenAI model \(GPT\-4o\)\. Applied advanced Prompt Engineering techniques, including multi\-turn conversational examples \(Few\-Shot Learning\) and strict system\-level instructions, to enforce a complex JSON output schema\. This increased data extraction accuracy and consistency by over 90% compared to basic prompting, virtually eliminating parsing failures\. 3\. Resilient & Autonomous Web Data Acquisition System \- Developed a resilient, multi\-source data acquisition pipeline using Python, Selenium, and BeautifulSoup\. Engineered a brand\-agnostic URL scoring algorithm to autonomously identify the most relevant datasheets from search engine results \(DuckDuckGo\)\. Successfully engineered solutions to overcome enterprise\-level challenges, including corporate network proxies, SSL interception, and anti\-bot mechanisms like Google's CAPTCHA\. 4\. Automation of High\-Value Business Processes 5\. Full\-Stack AI Project Implementation \- Owned the complete project lifecycle, from designing the modular Python architecture and Pydantic data models to implementing the final orchestration script \(main\.py\)\. Managed dependencies, configured environment variables \(\.env\), and established a robust logging and error\-handling framework, ensuring the tool is scalable, maintainable, and production\-ready\. Data Analyst/ Lead Data Engineer at Innovaccer, Oct 2024 \- Present · 1 yr 11 mos, Noida, Uttar Pradesh, India, Led a cross\-functional team and mentored 5\+ junior analysts to design scalable ETL pipelines, develop complex Snowflake SQL queries for L1\-to\-L2 data ingestion with 100% accuracy, and deliver Power BI dashboard for stakeholders Architected and implemented a scalable ETL pipeline to ingest 112M\+ rows of clinical data from ADLS, conducted data profiling using Python scripts,and automated workflows improving data processing time by 70% Automated Python pipeline for cleaning ,outlier treatment \(IQR\), and forecasting \(Prophet\) of acquisition counts and prices for 1,000\+ tech acquisitions; evaluated accuracy \(MAE, RMSE, MAPE\) and delivered dynamic Tableau dashboard Used MLflow to automate experiment tracking, parameter tuning, metric logging, and model versioning in a time series forecasting pipeline using Prophet, enabling reproducibility and deployment\-readiness of trained models Software Engineer\-BI/ Data Engineer at BOLD, Dec 2023 \- Sep 2024 · 10 mos, Noida, Uttar Pradesh, India, Developed the Transaction classification dashboard in MicroStrategy \(MSTR\) to display all transaction types in a single place\.  Collated multiple Sisense dashboards into a single one to enhance performance\.  Modified SQL queries to optimize cube performance and reduce data from 65 M rows to 37 M rows  Worked on Google AdWords alerts to identify campaigns responsible for conversion variance alerts\.  Performed QA for Google Sheet reports to validate functionality, data types, filters, SQL query logic, and terminology\. Optimized SQL queries to reduce intelligent cube dataset size from 65M to 37M rows, enhancing Microstrategy report performance and decreasing data retrieval time by 43% Developed advanced sequel queries in Snowflake to forecast user engagement metrics based on monthly budget allocations, improving forecasting accuracy by 25% and enabling data\-driven marketing decisions Business Analyst/ Big Data Engineer at EXL, Dec 2021 \- Dec 2023 · 2 yrs 1 mo, Remote, Project Name:\- KPI Dashboard • Analyzed the Customers Tables from Oracle and SQL server Database for US North and Texas region customers to make Data Dictionary \(Primary Key, Foreign Key, Key Columns and their description \) • Developed Exception Metric SQL code for KPI Dashboard that gives the details like Total Cases, Resolved Cases, Within SLA %, After SLA % for each brand , territory, region ,commodity\. • Responsible for Monthly Refresh of KPI Dashboard and perform Quality checks on data like Accuracy, Relevancy, Completeness, Timeliness • Modified SQL Scripts according to Ad\-hoc requirement from Clients like adding New Brands, combining two columns, extra filter condition, Removing Duplicates, etc\. Project Name:\- E\-Bill Health Check • Developed E\-Bill PBI Dashboard to drill\-down into customer E\-Bill operational Processes and end\-to\-end E\-bill Tracking • Created Summary Page to show E\-bill sent, E\-bill Not sent, Blank Emails, Bounce etc\. count on Card vis\. • Added drilldown pages for metrics mentioned on Summary page and applied page navigation with a focus on usability and performance\. • Improved E\-Bill Health by 40% by implementing interactive visualizations for Client • Used Visualization such as Tables, Cards, Gauge chart, Line chart etc\. to make report interactive and simple Project Name: Cash Outflow Forecasting • Forecasted Total Cash Outflow of Client using Python and Machine Learning Techniques • Performed EDA using PBI, Python, SQL & Excel and Deep\-dived into data to understand pattern, consistency, Outliers in payments\. • Collaborated with cross\-functional teams and researched on various Time Series models like ARIMA, XGBoost, FbProphet • Observed Pros and Cons of each ML technique and selected the best fitting one \(FbProphet\) • Developed PBI Dashboard to show Actual amount, Forecasted amount, MAPE values and received positive client feedback on personal approach and dedication\. Analyst/ Lead Data Analyst at AIDA Management, Feb 2020 \- Sep 2021 · 1 yr 8 mos, Noida, Uttar Pradesh, India, • Analyzing Business\-critical data to identify inefficiencies and trends in project • Monitor Project progress for Time, Cost and Quality through Monthly Progress report • Updating Project Documentation on a consistent basis and conveying them to the management • Coordinating with Client, Architect and Senior Management for requirement gathering and Quality checks • Able to Reduce the Cost of Project by 0\.5% by through Cost cutting of Construction material ## Education Bachelor of Technology \(B\.Tech\.\), Civil Engineering, Delhi Technological University \(Formerly DCE\) ## Public links - github: https://github\.com/ShubhamAIML\-Engineer [View the original profile](https://tal.club/meet/shubham-goyal) ```json { "@context": "https://schema.org", "@graph": [ { "@type": "Person", "@id": "https://tal.club/meet/shubham-goyal#person", "name": "Shubham Goyal", "url": "https://tal.club/meet/shubham-goyal", "image": "https://tal.club/meet/shubham-goyal/profile-photo.jpg", "description": "Meet Shubham Goyal, Enterprise GenAI & Big Data Architect | LLM Systems & RAG Architect | Data Engineer| Building Scalable GenAI Systems (2.5M+ tokens/day, <​50ms latency | Capgemini & Cisco 2026 AI excellence award winner based in Noida. AI stack includes antigravity and opencode. Explore public w...", "jobTitle": "Manager / AI architect ( Data & Infra)", "worksFor": { "@type": "Organization", "name": "Capgemini" }, "hasOccupation": { "@type": "Occupation", "name": "Manager / AI architect ( Data & Infra)" }, "homeLocation": { "@type": "Place", "name": "Noida" }, "alumniOf": [ { "@type": "EducationalOrganization", "name": "Delhi Technological University (Formerly DCE)" } ], "knowsAbout": [ "antigravity", "opencode" ], "sameAs": [ "https://github.com/ShubhamAIML-Engineer" ], "mainEntityOfPage": { "@id": "https://tal.club/meet/shubham-goyal#webpage" } }, { "@type": "ProfilePage", "@id": "https://tal.club/meet/shubham-goyal#webpage", "url": "https://tal.club/meet/shubham-goyal", "name": "Shubham Goyal on Tal", "description": "Meet Shubham Goyal, Enterprise GenAI & Big Data Architect | LLM Systems & RAG Architect | Data Engineer| Building Scalable GenAI Systems (2.5M+ tokens/day, <​50ms latency | Capgemini & Cisco 2026 AI excellence award winner based in Noida. AI stack includes antigravity and opencode. Explore public w...", "dateModified": "2026-08-02T06:51:28.851Z", "isPartOf": { "@id": "https://tal.club/#website" }, "mainEntity": { "@id": "https://tal.club/meet/shubham-goyal#person" }, "primaryImageOfPage": { "@type": "ImageObject", "url": "https://tal.club/meet/shubham-goyal/profile-photo.jpg", "contentUrl": "https://tal.club/meet/shubham-goyal/profile-photo.jpg", "caption": "Shubham Goyal profile photo on Tal" }, "inLanguage": "en-IN" } ] } ```