Apr 2024 – Present · 2 yrs 2 mos · India · Remote

• I mentor software engineers and aspiring developers through targeted 1:1 sessions on Topmate (Top 1% mentor) - practical, no-fluff guidance that accelerates your path to an offer.
• What you’ll get:
• Mock interviews with FAANG/top-tier rigor (DSA + System Design)
• Deep dives into HLD/LLD and structured problem-solving drills
• Resume & LinkedIn teardown with concrete rewrites and positioning
• Career strategy plus detailed written feedback and session summaries
• Why me:
• Senior SWE at LinkedIn (ex-Salesforce) with real production experience
• Recognized in the Top 1% on Topmate, rated 5/5 by mentees
• Have secured offers at Amazon, Microsoft, Adobe, Oracle, Tekion, and more
• Ready to step up your prep and confidence? Book a session → topmate.io/gauravkabra

May 2023 – May 2025 · 2 yrs · Hyderabad, Telangana, India · Hybrid

• Data Cloud (AKA Customer Data Platform, CDP and Genie) – Salesforce’s real-time hyperscale data engine – unifies and harmonizes customer data, enterprise content, telemetry data, Slack conversations, and other structured and unstructured data to create a single view of the customer. The platform is already processing 30 trillion transactions per month, and connecting and unifying 100 billion records every day.
• Dataspace support in Profile Explorer app. See https://tinyurl.com/profile-explorer
• Support new data-types: Phone, URL, Email and Percent. Salesforce’s native platform supports many complex data types such as currency but in Data Cloud, the platform only understood 4 data-types: Text, Number, Date and DateTime. Worked on supporting new data-types in 2 features: Data Transform (data prep recipe editor in Tableau CRM) & Data Actions. See https://tinyurl.com/data-prep-recipe & https://tinyurl.com/data-actions-and-targets
• Actively worked customer investigations (including Sev1) in compliance with the SLAs
• Worked on company-wide high priority Gen AI foundational unstructured data service from scratch - retrieval engine, specifically semantic search, based on chunking and vector embedding to deliver trusted, coherent, contextual, and personalized responses grounded to Salesforce’s customers’ metadata and data
• This generative AI feature will support any type of data including text, image, spatial and unstructured data and expandable to retrieval augmented generation (RAG) for grounding LLMs
• This feature leverages Microsoft E5 embedding model, Milvus DB, Trino, Kafka, S3, EMR on EKS, DynamoDB, RDS, Apache Iceberg, Parquet and more cutting-edge technology

May 2022 – Apr 2023 · 11 mos · Hyderabad, Telangana, India · Hybrid

• Connect API enhancements: new attributes and record type changes in newer versions of IDR APIs. Connect is Salesforce's framework for developing APIs
• Drove the entire discussions with the US Health Cloud team regarding lock-in plan for record type changes
• Led the discussion for UI modularity for new components in IDR's UX Lo-Fi mocks
• Data model changes for table support feature
• Implemented the transaction logic in Intelligent Document Reader (IDR) - a long pending item under the Trust & Tech Debt bucket: When user saves template, data is stored in lot of entities. If some exception occurs in mid-way of saving, database will be in inconsistent state. Remediated the problem with the implementation
• Enabled try-before-buy onboarding via scratch org support in IDR, accelerating customer acquisition
• POCs on key-phrase extraction using AWS Comprehend, Query support using AWS Textract, and Salesforce Community support
• Delivered the 3 invocable actions - the top asks from the Financial Services Cloud. Was appreciated by the GM & EVP of the cloud, PM, and managers alike. The actions are pluggable in Salesforce Flows on UI, Apex in Developer Console, exposed as REST endpoints, and available in OmniScript
• Metadata support for newer fields in IDR data model. Salesforce customers rely a lot on metadata to build their custom applications and migrate the data from one org to another
• Perf improvement in saving logic of IDR templates of 11%
• Explored the Java Lucene library for NLP and Apache Tika for language detection to explore if dependency on AWS can be eliminated
• Worked on investigations from customers like Eli Lilly and Cloudforia and patched back proper fixes within timelines. The investigation from Eli Lilly was subtle and in tight SLA. I found that the Java class mostly used by engineers for querying DB was not actually meant to be used in the UI-tier. I used and tested other classes thoroughly and unblocked the customer

Jun 2020 – Apr 2022 · 1 yr 10 mos · Hyderabad, Telangana, India · Hybrid

• Industries Cloud offers tailored solutions for diverse sectors like healthcare, finance, manufacturing, and communications, addressing their distinct requirements through industry-specific offerings.
• Worked in two teams under the umbrella of Industries Einstein - the Metamind team and the Einstein Platform (EP) team.
• In the EP team,
• worked on the CRM Analytics platform (FKA Einstein Discovery or Tableau CRM platform)
• created training set recipes (the ELT framework of CRM Analytics) and UI enhancements for Visit Recommendation product
• In the Metamind team,
• worked on the Intelligent Form Reader (IFR) product (based on AWS Textract). IFR is consumed by verticals like Health Cloud (HC), Financial Service Cloud (FSC) and Public Sector & has customers such as Johnson & Johnson and Ministry of Health, British Columbia
• worked on the Sentiment Insights (FKA Sentiment Analysis) product (based on AWS Comprehend). Sentiment Insights is consumed by the Surveys team to detect polarity and emotions of customers
• enterprise projects die at the compliance stage - made IDR data-residency compliant for sell in Europe, Canada and Australia
• Due to this work, Salesforce could onboard customers in these countries - one of the notable examples is the Health Ministry of British Columbia
• developed Connect APIs for consuming clouds (Connect is Salesforce's framework to develop APIs)
• fixed bugs and dependency test failures in Trust & Tech Debt
• conduct spikes (POCs)
• Languages/Tools/Frameworks: Java, JavaScript, AWS APIs, Lightning Web Components (LWC), Aura Components, Apex, CRM Core Platform, CRM Analytics/Tableau CRM, JSON, IntelliJ IDEA, VS Code, Slack APIs, Heroku, Eclipse

May 2019 – Jul 2019 · 2 mos · Chennai, Tamil Nadu, India · On-site

• Engaged as an intern on the impactful project "OBI Server Log Analyzer & Health Prediction Using Machine Learning," where I played a pivotal role in predicting the health status of the Oracle Business Intelligence (OBI) server.
• Key Achievements:
• Conducted extensive testing of various machine learning models, including Support Vector Machines (SVMs), Decision Trees, and ensemble learning methods such as Random Forests, to determine the most effective solution
• Employed rigorous evaluation metrics to fine-tune and optimize the selected model for accurate health predictions
• Technical Skills:
• Utilizing a range of tools, and frameworks, including Numpy, Pandas, Matplotlib, Seaborn, Re, Tkinter, Win32com.client, Graphviz, Jupyter Notebook with Conda environment, and PyCharm Community Edition
• Contributions:
• Demonstrated expertise in data analysis and visualization, ensuring a comprehensive understanding of the OBI server's health
• Successfully implemented the chosen machine learning model, enhancing the overall efficiency and reliability of health predictions for the OBI server