Articles by Sumit
Activity
47K followers
-
Sumit Gupta shared thisIf I had to start Data Engineering from zero again, this is the exact roadmap I would follow. To master Data Engineering, the order matters more than most beginners realize. I see many people start with SQL one week, Python the next, then suddenly jump into Kafka, cloud, Airflow, or some random tool they saw online. That is where the confusion starts. Not because they are bad learners, but because no one showed them how these pieces fit together in a real data pipeline. So instead of learning everything randomly, here is the roadmap I would follow today, with free resources: 1. SQL Start with joins, CTEs, aggregations, window functions, subqueries, and query logic. Free resource: SQL Full Course by freeCodeCamp https://lnkd.in/gMDj8EGh Practice: https://sqlbolt.com 2. Python Learn enough Python to clean files, work with APIs, automate tasks, and handle data. Free resource: Python Full Course by freeCodeCamp https://lnkd.in/gzzE5DVr Practice: https://lnkd.in/gitsS4cU 3. Databases Understand tables, keys, indexes, normalization, transactions, and PostgreSQL basics. Free resource: PostgreSQL Tutorial https://lnkd.in/gvMrzS63 Channel: https://lnkd.in/gt6ThdAa 4. Data Engineering Basics Learn how data moves from sources into storage, warehouses, and pipelines. Free resource: Data Engineering Zoomcamp https://lnkd.in/grVZS2Px Channel: https://lnkd.in/gxadfN3u 5. Data Warehousing Learn facts, dimensions, star schema, partitioning, clustering, and data modeling. Free resource: https://lnkd.in/gnd2M88v 6. dbt Learn SQL transformations, testing, documentation, lineage, and analytics engineering workflows. Free resource: https://lnkd.in/gPQj636c Channel: https://lnkd.in/g5nRCK4s 7. Airflow Learn DAGs, scheduling, dependencies, retries, and pipeline orchestration. Free resource: Airflow 101 by Astronomer https://lnkd.in/grBHPkSm Channel: https://lnkd.in/gAVJt5vq 8. Kafka Learn events, topics, producers, consumers, and streaming data basics. Free resource: Kafka 101 by Confluent https://lnkd.in/gnZWsnyN Channel: https://lnkd.in/gy9BrHzj 9. Projects Build an API pipeline, batch ETL pipeline, dbt model, Airflow DAG, warehouse model, and Kafka mini-project. Because courses teach concepts. But projects create proof. And in Data Engineering, proof beats certificates.
-
Sumit Gupta shared thisMy friend was about to leave the US. Staff engineer. 12 years of experience. Built systems processing millions of transactions daily. But his H-1B transfer was stuck. His wife's EAD renewal — pending 11 months. He turned down a Staff role at a startup because the visa risk was too high. He told me: "I've given this country nine years of my best work. And I still can't plan six months ahead." Then I asked him one question: "Have you looked into the O-1A?" He laughed. "That's for famous people." It's not. I walked him through it: → That distributed system you built that saved your company $3M? Original contribution of major significance. → That conference talk at KubeCon? Evidence of your work being showcased at significant events. → Your base comp in the top 5% for your role? High salary relative to others in the field. His face changed. He had FOUR criteria sitting in his LinkedIn profile. He just didn't know they counted. The O-1A doesn't require employer sponsorship the way H-1B does. It's not tied to a lottery. And the bar is high — but not as high as most people think. The real barrier isn't qualification. It's information. My friend isn't leaving the country anymore. He's building his petition. If you're a tech professional stuck in H-1B limbo, you might have more options than you realize. Check out Manifest Law (Manifest Law) — They offer free consultation if eligible, they'll tell you if you already qualify for O1A if you complete the quiz. Sign up using the link : https://lnkd.in/gVMSt-pB Know someone who needs to see this? Send it to them. 👇
-
Sumit Gupta shared thisBad data breaks good systems. Dashboards, ML models, and analytics pipelines are only as reliable as the data feeding them. Before trusting any dataset, it’s important to run a few basic data quality checks. Here are 10 SQL data quality checks every data team should run. - NULL Check Detect missing values in important columns like IDs, dates, or keys. Missing values often indicate incomplete ingestion or pipeline failures. - Uniqueness Check Ensure primary keys or unique identifiers are not duplicated. Duplicate IDs can corrupt aggregations and downstream joins. - Referential Integrity Check Verify relationships between tables. Every foreign key should match a valid record in the parent table. - Range Validation Confirm numeric values fall within expected limits. Out-of-range values often signal data entry errors or pipeline bugs. - Accepted Values Check Validate that categorical fields contain only allowed values. Unexpected categories can break business logic or reporting. - Data Type Validation Ensure fields match the expected data type. Incorrect types can cause query failures or inaccurate transformations. - Date Consistency Check Validate chronological logic. For example, an end date should never occur before a start date. - Data Freshness Check Confirm that pipelines are loading recent data. Outdated data often indicates broken ingestion jobs. - Outlier Detection Identify unusual spikes or anomalies in metrics such as revenue, transactions, or event counts. - Volume Check Monitor record counts to detect sudden drops or unexpected spikes in incoming data. Data teams often focus on building pipelines. But reliable systems come from verifying data quality before the data is used anywhere else. Which data quality check do you run first before trusting a new dataset? Follow Sumit Gupta for more such insights!!
-
Sumit Gupta shared thisAI learning explained in one meme: Human: “Paint the fence like this.” AI: “Understood. I have detected the pattern.” Also AI after 5 minutes: “I painted the fence based on statistical confidence, not common sense.” That’s the funny part about AI. It does not always understand the goal. It learns from examples, spots patterns, and predicts what should come next. Sometimes that feels magical. Sometimes it gives you a fence that looks like a corrupted barcode.
-
Sumit Gupta reposted thisYour resume gets scanned in 2 minutes. Your portfolio gets you hired. Most data professionals spend weeks polishing a resume, then send it into a black hole. The truth? Hiring managers don't want a list of skills. They want proof you can solve real problems with real data. That's what a strong portfolio does. Here’s how to build one: ✅ Start with your target role - stop building random projects ✅ Pick 3-4 deep projects over 10 shallow ones ✅ Show end-to-end thinking, not just clean charts ✅ Add business storytelling (what happened, why it matters, what to do) ✅ Choose the right platform - GitHub for code, a website for presentation ✅ Put it everywhere: resume, LinkedIn Featured, cover letters One complete project beats five half-finished ones. Your portfolio is nott a folder of projects. It is proof of your skill, thinking, and business value. Swipe through 👉 and save it for your next job hunt. Follow Sumit Gupta for more such insights!!
-
Sumit Gupta shared thisYour resume gets scanned in 2 minutes. Your portfolio gets you hired. Most data professionals spend weeks polishing a resume, then send it into a black hole. The truth? Hiring managers don't want a list of skills. They want proof you can solve real problems with real data. That's what a strong portfolio does. Here’s how to build one: ✅ Start with your target role - stop building random projects ✅ Pick 3-4 deep projects over 10 shallow ones ✅ Show end-to-end thinking, not just clean charts ✅ Add business storytelling (what happened, why it matters, what to do) ✅ Choose the right platform - GitHub for code, a website for presentation ✅ Put it everywhere: resume, LinkedIn Featured, cover letters One complete project beats five half-finished ones. Your portfolio is nott a folder of projects. It is proof of your skill, thinking, and business value. Swipe through 👉 and save it for your next job hunt. Follow Sumit Gupta for more such insights!!
-
Sumit Gupta shared thisI was born in Dharavi, Mumbai. A place often described as one of the world’s largest slums. But for me, it was home. Growing up, even access to proper education was not easy, and the idea of building a tech career in Silicon Valley felt almost impossible. No elite-school network. No family blueprint. No clear path to follow. So when people now see: Dharavi → Superhuman India → California No connections → top tech companies It sounds like one big inspiring jump. But it was not. The middle was messy. It was years of learning things no one around me fully understood. Applying, getting ignored, trying again, and wondering if I was even moving in the right direction. No roadmap means you guess a lot. You fail quietly. And you doubt yourself far more than you ever admit publicly. Early in my career, I thought certifications would change everything. So I kept collecting them. → One more course. → One more badge. → One more line on the resume. But certificates alone did not change my life. What actually changed things was much less glamorous. Showing up every week. Breaking down data problems. Sharing what I was learning about analytics, dashboards, SQL, metrics, and the modern data stack - even when almost no one was watching. No big audience. No instant validation. No clear signal that it was working. Just repetition. Week after week. That is the part people skip. They see the outcome. They do not see the compounding. Because going from Dharavi to Silicon Valley was never one giant leap. It was hundreds of small, invisible steps stacked over years. And most of those steps felt like they were going nowhere. Until suddenly, they started opening doors I once thought were impossible.
-
Sumit Gupta reposted thisYou can learn Python Faster by building things with it. Not random syntax. Not endless theory. Actual projects that force you to solve real problems. That’s why I created this list of 15 Python projects you can build to move from beginner to advanced in your data career. Start with simple logic-based projects. Then move into automation, APIs, databases, AI, data pipelines, and full-stack applications. Here are the GitHub resources: 1. Interactive Word Game https://lnkd.in/guwUh7bj 2. Calculator App https://lnkd.in/gS3MSUtZ 3. To-Do List App https://lnkd.in/gHdgDZCk 4. Automating File Organization https://lnkd.in/g8Vfq5B4 5. Password Generator https://lnkd.in/g484M6eG 6. Food Ordering App https://lnkd.in/gNyYN5fr 7. Dynamic AI Chatbot https://lnkd.in/gzJzHZzS 8. Weather App with API https://lnkd.in/gx3qhQZT 9. Sending Automated Emails https://lnkd.in/gpttGgSk 10. Data Pipeline with APIs https://lnkd.in/g7yJfx3G 11. Analyzing Wikipedia Pages https://lnkd.in/gBCwVReE 12. Key-Value Database https://lnkd.in/g4aQWPgq 13. Speech Recognition & Summarization System https://lnkd.in/gKYKgHgE 14. Recommendation System https://lnkd.in/gHVn-tSm 15. Full-Stack Web App with Auth & APIs https://lnkd.in/g_EEegFs Small tip: Clone one repo. Run it locally. Change the logic. Add one extra feature. Write down what each file does. That is where Python starts becoming a real skill. ♻️ Repost this to help someone learning Python. ➕ Follow Sumit Gupta for more practical data and AI learning resources.
-
Sumit Gupta shared thisYou can learn Python Faster by building things with it. Not random syntax. Not endless theory. Actual projects that force you to solve real problems. That’s why I created this list of 15 Python projects you can build to move from beginner to advanced in your data career. Start with simple logic-based projects. Then move into automation, APIs, databases, AI, data pipelines, and full-stack applications. Here are the GitHub resources: 1. Interactive Word Game https://lnkd.in/guwUh7bj 2. Calculator App https://lnkd.in/gS3MSUtZ 3. To-Do List App https://lnkd.in/gHdgDZCk 4. Automating File Organization https://lnkd.in/g8Vfq5B4 5. Password Generator https://lnkd.in/g484M6eG 6. Food Ordering App https://lnkd.in/gNyYN5fr 7. Dynamic AI Chatbot https://lnkd.in/gzJzHZzS 8. Weather App with API https://lnkd.in/gx3qhQZT 9. Sending Automated Emails https://lnkd.in/gpttGgSk 10. Data Pipeline with APIs https://lnkd.in/g7yJfx3G 11. Analyzing Wikipedia Pages https://lnkd.in/gBCwVReE 12. Key-Value Database https://lnkd.in/g4aQWPgq 13. Speech Recognition & Summarization System https://lnkd.in/gKYKgHgE 14. Recommendation System https://lnkd.in/gHVn-tSm 15. Full-Stack Web App with Auth & APIs https://lnkd.in/g_EEegFs Small tip: Clone one repo. Run it locally. Change the logic. Add one extra feature. Write down what each file does. That is where Python starts becoming a real skill. ♻️ Repost this to help someone learning Python. ➕ Follow Sumit Gupta for more practical data and AI learning resources.
Sumit Gupta replied to a comment
18m
Poornachandra Kongara That was exactly my thinking. SQL is the language of data, and most engineering tasks eventually connect back to how data is stored, queried, transformed, or modeled. Building that foundation early reduces confusion when more advanced tools enter the picture.
Sumit Gupta replied to a comment
18m
ZEN EB5 Absolutely. Most resources teach tools in isolation, but real-world data engineering is about how those tools interact. Once people understand the flow of data through a pipeline, the learning process becomes much more structured.
Sumit Gupta replied to a comment
18m
Ananya Ghosh Chowdhury Very true. Those concepts appear repeatedly in both interviews and production environments. A strong SQL foundation pays dividends across analytics, engineering, and data platform roles.
-
Sumit Gupta liked thisSumit Gupta liked thisNothing beats the energy of connecting with an incredible DSH team in person at the Snowflake Summit! Ready for an amazing week of learning, networking, and driving innovation together. #SnowflakeSummit #DataSuperhero #RajivGupta
-
Sumit Gupta liked thisSumit Gupta liked thisWhat if the biggest obstacle to AI success isn't the technology? It's the way organizations are structured. At Data Citizens on the Road by Collibra, I sat down with Joyce Snelders Senior Manager at Deloitte on The Ravit Show to discuss what organizations are experiencing as they move from AI experimentation to enterprise-wide adoption. A few key takeaways from our conversation: • Data governance has gone from a "nice to have" to a business priority because AI is only as good as the data behind it. • Many organizations are building AI agents without common standards, creating duplicate efforts and inconsistent outcomes across teams. • Chief Data Officers are increasingly becoming AI leaders, taking responsibility for both data and AI strategies. • The next phase of enterprise AI is not just about technology. It is about governance, operating models, and change management. • Leaders should start preparing for a future where digital FTEs work alongside human employees. One statement from Joyce stood out: Organizations don't have an AI problem. They have a governance and operating model problem. The companies that solve that challenge first will be the ones that scale AI successfully. #DataCitizens #Collibra #AI #DataGovernance #AIGovernance #EnterpriseAI #DataLeadership #TheRavitShow
-
Sumit Gupta reacted on thisIf I had to start Data Engineering from zero again, this is the exact roadmap I would follow. To master Data Engineering, the order matters more than most beginners realize. I see many people start with SQL one week, Python the next, then suddenly jump into Kafka, cloud, Airflow, or some random tool they saw online. That is where the confusion starts. Not because they are bad learners, but because no one showed them how these pieces fit together in a real data pipeline. So instead of learning everything randomly, here is the roadmap I would follow today, with free resources: 1. SQL Start with joins, CTEs, aggregations, window functions, subqueries, and query logic. Free resource: SQL Full Course by freeCodeCamp https://lnkd.in/gMDj8EGh Practice: https://sqlbolt.com 2. Python Learn enough Python to clean files, work with APIs, automate tasks, and handle data. Free resource: Python Full Course by freeCodeCamp https://lnkd.in/gzzE5DVr Practice: https://lnkd.in/gitsS4cU 3. Databases Understand tables, keys, indexes, normalization, transactions, and PostgreSQL basics. Free resource: PostgreSQL Tutorial https://lnkd.in/gvMrzS63 Channel: https://lnkd.in/gt6ThdAa 4. Data Engineering Basics Learn how data moves from sources into storage, warehouses, and pipelines. Free resource: Data Engineering Zoomcamp https://lnkd.in/grVZS2Px Channel: https://lnkd.in/gxadfN3u 5. Data Warehousing Learn facts, dimensions, star schema, partitioning, clustering, and data modeling. Free resource: https://lnkd.in/gnd2M88v 6. dbt Learn SQL transformations, testing, documentation, lineage, and analytics engineering workflows. Free resource: https://lnkd.in/gPQj636c Channel: https://lnkd.in/g5nRCK4s 7. Airflow Learn DAGs, scheduling, dependencies, retries, and pipeline orchestration. Free resource: Airflow 101 by Astronomer https://lnkd.in/grBHPkSm Channel: https://lnkd.in/gAVJt5vq 8. Kafka Learn events, topics, producers, consumers, and streaming data basics. Free resource: Kafka 101 by Confluent https://lnkd.in/gnZWsnyN Channel: https://lnkd.in/gy9BrHzj 9. Projects Build an API pipeline, batch ETL pipeline, dbt model, Airflow DAG, warehouse model, and Kafka mini-project. Because courses teach concepts. But projects create proof. And in Data Engineering, proof beats certificates.
-
Sumit Gupta liked thisSumit Gupta liked thisA "busy" EB-1A file and an "extraordinary" one can list the same achievements and get opposite outcomes. The difference isn't how much you've done. It's how it reads to an officer. Here's the contrast, dimension by dimension 👇 → Core narrative - busy: many activities, no through-line. Extraordinary: one clear theme ties it all together. → Impact - busy: worked on many things, outcomes vague. Extraordinary: clear, measurable influence beyond titles. → Awards - busy: several, but low-signal. Extraordinary: fewer, but prestigious and selective. → Media - busy: self-driven or weak outlets. Extraordinary: independent, credible coverage. → Leadership - busy: impressive titles, unclear scope. Extraordinary: defined by decisions and outcomes. → Original contributions - busy: useful but not novel. Extraordinary: new, adopted, cited, widely used. → Recommendation letters - busy: heavy praise, little substance. Extraordinary: specific impact and genuine respect. → Consistency - busy: documents tell different stories. Extraordinary: every exhibit reinforces one identity. → Quality vs quantity - busy: large but repetitive. Extraordinary: curated, each piece adds value. Officer's takeaway? Busy reads "capable." Extraordinary reads "top of field." Save this before you build your file. Which column does your profile fall into right now? 👇 Want to learn more about O1, EB1A and EB5? Schedule a free consultation - https://lnkd.in/gEf3y7JB Join our upcoming free webinars with experts across immigration, careers, and admissions - https://luma.com/qyhm0fug Meet sponsor-friendly employers, investors, and attorneys in person in the Bay Area this August - https://luma.com/oas2026 🔔 Follow to stay updated on high-skilled immigration, jobs, and tech #H1B #ImmigrationJourney #GreenCard #EB1A #EB5 #USImmigration
-
Sumit Gupta liked thisSumit Gupta liked thisJune 1. Last month of Q2. Whatever you set out to do this quarter — you still have 30 days to make it real. What's the one thing you're finishing strong?
-
Sumit Gupta liked thisSumit Gupta liked thisYour AI can be perfectly safe and still get you sued. It can pass every audit and still make terrible decisions. It can be "well governed" and still hallucinate in production. Governance, safety, and compliance are NOT the same thing and treating them as one is how AI programs quietly fall apart. 👇 → GOVERNANCE = the WHY. The org-wide framework: strategic alignment, accountability, transparency, lifecycle oversight. → Powerful AI that stays controllable. → SAFETY = the BEHAVIOR. Keeps AI reliable and harmless: behavioral control, alignment safeguards, hallucination reduction, fail-safes, human-in-the-loop. → Protects users and outcomes. → COMPLIANCE = the RULES. Meets laws and standards: regulatory adherence, data privacy, audit readiness, consent, reporting. → Prevents legal risk. Governance sets direction. Safety earns trust. Compliance keeps you legal. Miss one and the other two can't save you. Save this for your AI strategy. Which one is your org weakest on? 👇 Want to learn more about O1, EB1A and EB5? Schedule a free consultation - https://lnkd.in/gEf3y7JB Join our upcoming free webinars with experts across immigration, careers, and admissions - https://luma.com/qyhm0fug Meet sponsor-friendly employers, investors, and attorneys in person in the Bay Area this August - https://luma.com/oas2026 🔔 Follow to stay updated on high-skilled immigration, jobs, and tech
-
Sumit Gupta liked thisSumit Gupta liked this"They were so interested last week." Now they won't reply. Sound familiar? Every recruiter has watched a hot candidate go silent and wondered what changed. Usually, nothing changed on their side - everything changed (or didn't move) on yours. Here are the top 7 reasons candidates ghost recruiters 👇 1. Outreach Comes Too Late - top candidates are already in another process 2. Messages Feel Too Generic - vague outreach gets ignored 3. Follow-Ups Are Inconsistent - one touchpoint is rarely enough 4. The Role Is Unclear - candidates lose interest when expectations are fuzzy 5. The Process Feels Too Slow - long silences create doubt 6. Candidates Are Not Properly Qualified - the role was never a real fit 7. Recruiters Are Overloaded - manual workloads guarantee delays Ghosting isn't random. It's a pattern. And patterns can be fixed. 💡 I've created a free detailed AI Hiring Playbook to help HR leaders and founders move from Manual → Agentic AI Hiring - so you stop burning hours on manual tasks and start building a hiring engine that runs itself. 1. Like this post ❤️ 2. Comment "AI Hiring" below 3. Follow me for more HR & hiring insights 4. I'll send it directly to your DMs 👉 Which reason hurts your team most? Drop the number below. 👉 Try Catch.hr: www.catch.hr 👉 Book a free demo with me: https://lnkd.in/dzzpcPAb
Experience & Education
-
Superhuman
View Sumit’s full experience
See their title, tenure and more.
or
Licenses & Certifications
-
Certified Solutions Architect - Associate
Amazon Web Services (AWS)
IssuedCredential ID AWS00386280 -
-
-
-
-
-
-
-
-
Volunteer Experience
-
Digital Media Voluteer
Nanhikali
- Present 10 years 10 months
Children
Publications
-
1-Click Marketing Blueprint Formulation using Web Mining
International Journal of Engineering Research and Technology
This paper proposes an idea in which a marketing application will help us to create the whole marketing blueprint using just a click. With the evolution of the world, the success of the product now not only depends on the quality of the product but also depends on how well the company is able to market their products and bring it to the masses. The day has arrived wherein the public will know about the product when it is marketed in a way that it reaches the masses and is accessible to the…
This paper proposes an idea in which a marketing application will help us to create the whole marketing blueprint using just a click. With the evolution of the world, the success of the product now not only depends on the quality of the product but also depends on how well the company is able to market their products and bring it to the masses. The day has arrived wherein the public will know about the product when it is marketed in a way that it reaches the masses and is accessible to the targeted customer. But the issue with marketing is that we need to market research before formulating a plan and implementing it. In this system, we will create a marketing blueprint with just couple of clicks and present a proper strategy as the output which will be categorized based on the industry, budget and type of marketing(online and offline).
Other authors -
-
Semantic Web: using Microdata and Schema
International Conference and Workshop on Advance Computing
Other authors -
Courses
-
Applied Data Science
IST 687
-
Business Analytics
SCM 600
-
Database Management System
IST 659
-
Enterprise Risk Management
IST 625
-
IT Auditing
IST 600
-
Information System Analysis
IST 654
-
Introduction to Information Management
IST 621
-
Leadership and Cultural Exchange
-
Projects
-
Data Warehousing: NOPCommerce Enterprise Data Warehouse
♦ Conducted ETL with SQL SSIS & MOLAP Cube Development using SQL SSAS from relational star schema and created data pipelines from source to the Enterprise Data Warehouse
♦ Designed dimensional model for nop-commerce database using High level dimensional modeling and detailed level dimensional modeling techniques.
♦ Performed the analysis of sales figure for an internally created e-commerce portal & proposed methods of increasing sales
♦ Visualized the data using PowerBI and Tableau…♦ Conducted ETL with SQL SSIS & MOLAP Cube Development using SQL SSAS from relational star schema and created data pipelines from source to the Enterprise Data Warehouse
♦ Designed dimensional model for nop-commerce database using High level dimensional modeling and detailed level dimensional modeling techniques.
♦ Performed the analysis of sales figure for an internally created e-commerce portal & proposed methods of increasing sales
♦ Visualized the data using PowerBI and Tableau and presented the findings to increase and sales and e-commerce site traffic
Other creators -
-
Enterprise Risk Management - NEXIS Lab
- Present
♦ Assessed the risk culture and risk infrastructure of an organization by carrying out in detailed analyses using Surveys, interviews & SWOT analysis
♦ Conceptualizing an enterprise risk process model by identifying different risks for the lab, Measuring those risks based on probability of occurrence and impact
♦ Prioritizing and categorizing different risks based on the impact and delineating possible mitigation techniques for the sameOther creators -
-
Data Analysis: U.S. Department of Agriculture
-
♦ Conducted exploratory data analysis of U.S. Department of Agriculture dataset containing health and restaurant statistics in U.S.
♦ Performed data cleaning in R by removing inconsistencies in data such as missing values and incorrect entries
♦ Created visualizations (using ggplot2, chloropleth package)Other creators -
-
Business Analytics: Recruitment Strategy and Loan Application Approval
-
♦ Data Analysis using Tableau & Google Analytics. Used analytical techniques such as PowerPivot and MS Excel to visualize data patterns to strategizing cities for the recruitment tour. Applied Cost OPTIMIZATION techniques to reduce the campaigning costs by 10% by strategizing utilization of resources.
♦ Performed Statistical analysis on loan distribution data using Linear, Logit & Probit Regression models in R-Solver and R-Commander.
♦ Used sensitivity analysis in excel on the factors…♦ Data Analysis using Tableau & Google Analytics. Used analytical techniques such as PowerPivot and MS Excel to visualize data patterns to strategizing cities for the recruitment tour. Applied Cost OPTIMIZATION techniques to reduce the campaigning costs by 10% by strategizing utilization of resources.
♦ Performed Statistical analysis on loan distribution data using Linear, Logit & Probit Regression models in R-Solver and R-Commander.
♦ Used sensitivity analysis in excel on the factors identified to generate patterns which would increase the probability of success of a loan application
♦ Performed Sensitivity Analysis, Predictive Modelling and Price Optimization using Excel (Pivot Tables, Power Pivot, Data Mining, Power View and Regression)
Other creators -
Honors & Awards
-
Best Photographer of the year
Thakur College of Engineering and Technology
Languages
-
English
-
-
Hindi
-
-
Marathi
-
Organizations
-
Nanhi Kali: Girl Child Education
-
- Present -
Institute of Electrical and Electronics Engineers
-
Recommendations received
2 people have recommended Sumit
Join now to viewOther similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content