Kumar Saksham.
DATA · ANALYTICS · RISK · PRODUCT

Turning Complex Data into Clear, Defensible Decisions.

Engineering graduate focused on analytics, risk, customer insights, and AI-assisted decision systems.

View My Work
GitHubLinkedInkumarsaksham560@gmail.com
SCROLL TO EXPLORE
CORE DOMAINS

Analytics · Data · Risk · Product

AnalyticsCustomer insights, segmentation and statistical analysis.
DataReliable pipelines, SQL and data quality.
RiskAnomaly detection, scoring and validation.
ProductProblem framing, prioritization and decision support.
Analytical Foundation · Decision Focus

Analytics in Service of Better Decisions

I combine structured analysis, statistical methods, and reliable data workflows to solve customer, risk, and data-quality problems and support better decisions.

My academic foundation is in Electronics and Communication Engineering at BIT Mesra, complemented by research at LNMIIT. Working across electromagnetic simulation, signal processing, and numerical optimization taught me to approach complex technical problems from first principles.

I now apply that structured mindset to Customer Analytics, Risk Modeling, and Data Systems. My work focuses on evaluating real data scale, uncovering verifiable signals, and building reliable decision-support systems.

Customer & Product Analytics

Customer behavior, segmentation, opportunity modeling, and decision support.

Risk & Statistical Analysis

Risk scoring, anomaly detection, statistical validation, and evidence-based analysis.

AI-Assisted Analytics

LLM APIs, structured AI outputs, risk narratives, and AI-assisted analytical workflows.

Data & Analytical Systems

SQL, data quality, reliable workflows, dashboards, and analytical implementation.

Evidence & Scale

Audited Portfolio Telemetry & Impact

Real dataset scale analyzed across forensic invoice audit, regulatory data reliability pipelines, and multi-million transaction retail lakes.

0K+
Invoice Records
GST Forensic Audit
0.0M+
Transactions
Customer Segmentation
$0.00M
Sales Analyzed
Promotional Uplift
0K+
E-Way Bill Records
Data Reliability Pipeline
Professional Experience

Research & Industry Experience

Rigorous hands-on algorithmic problem solving, electromagnetic simulation, and quantitative research.

Business Analysis & Operations

Business Analyst Intern

Sun Shine Automation India Ltd
Jan 2026 – June 2026Jaipur, India
Analyzed historical B2B order data, seasonal demand patterns, and after-sales service records using Excel, SQL, and Python. Identified recurring client requirements to support demand forecasting, proactive production planning, and maintenance scheduling, contributing to improved planning efficiency and engineering resource utilization.
Key Contributions & Focus Areas
DEMAND FORECASTING & TREND ANALYSIS01
Analyzed historical B2B order data and seasonal demand trends using Excel and SQL to identify recurring purchasing patterns and anticipate future client requirements.
PRODUCTION PLANNING OPTIMIZATION02
Leveraged historical order patterns to anticipate repeat orders and align production schedules with projected client demand, contributing to an approximately 18% improvement in production planning efficiency.
AFTER-SALES SERVICE & RESOURCE PLANNING03
Analyzed historical maintenance and service requirements to anticipate after-sales demand, enabling better engineer scheduling and workforce utilization.
Key Highlights & Achievements
  • Applied Excel, SQL, and Python to analyze historical order data and identify recurring B2B purchasing patterns.
  • Supported proactive production scheduling through demand forecasting, contributing to an approximately 18% improvement in production planning efficiency.
  • Used historical maintenance trends to anticipate service requirements and improve engineering resource allocation.
Key Outcomes & Metrics
0
Planning Efficiency Improvement
Historical Order Analysis
0
Analytical Tools Used
Excel, SQL & Python
0
Operational Areas Analyzed
Production & After-Sales Service
Technologies & Tools:Microsoft ExcelSQLPythonHistorical Data AnalysisTrend AnalysisDemand ForecastingProduction PlanningWorkforce Planning
LUSIP 2025 Research Fellowship

Summer Research Intern

The LNM Institute of Information Technology (LNMIIT)
May 2025 – July 2025Jaipur, India
Investigated 3D chipless RFID structures through electromagnetic simulation, MATLAB-based data processing, and experimental validation, focusing on improving tag performance and extracting reliable resonant signatures.
Research Areas & Methodology
DATA PROCESSING01
Processed 20K+ RFID signatures through a modular MATLAB workflow, converting simulation outputs into structured data for analysis.
CONFIGURATION ANALYSIS02
Compared four resonator configurations and multiple feed variations to identify patterns associated with improved chipless RFID performance.
VALIDATION03
Cross-checked experimental results against simulation outputs, investigated discrepancies, and refined a reproducible analytical workflow.
Key Highlights & Achievements
  • Engineered a modular MATLAB data-processing workflow to automate processing and analysis of 20K+ RFID signatures.
  • Analyzed simulation data across four resonator configurations and multiple feed variations to identify optimal tag performance.
  • Cross-checked experimental results against simulation outputs, investigating discrepancies to document a reproducible workflow.
  • Contributed to measured research improvement in tag utilization efficiency from 66% to 80%.
Research Outcomes & Metrics
0
RFID Signatures Processed
Converted simulation outputs into structured datasets
0
Tag Utilization Efficiency
Measured technical performance improvement
0
Resonator Configurations
Multi-geometry analysis across feed variations
Technologies & Tools:CST Studio SuiteMATLABElectromagneticsOptimization AlgorithmsSignal ProcessingQuantitative R&DSystems Modeling
Flagship Projects

Work & Flagship Projects

Evidence-based projects supporting customer opportunity modeling, forensic anomaly detection, regulatory data reliability, and payment uncertainty recovery.

Customer & Product AnalyticsCustomer Analytics

Promotional Opportunity & Customer Segmentation Analysis

Customer segmentation, basket affinity, and household promotional opportunity scoring in Power BI

Analyzed 2.6M+ transactions across 2,500 households representing $8.06M in sales to understand customer behavior and identify promotional opportunities.

Transactions
2.6M+
Households
2,500
Sales Analyzed
$8.06M
Technical Highlights
  • Engineered a household-level promotional scoring framework evaluating customer engagement, category affinity, and campaign responsiveness.
  • Controlled for analytical zero-inflation bias and department-size distortion across 2,500 households and $8.06M in sales.
  • Translated and cross-validated analytical pipelines across Python, PostgreSQL, and AWS Athena.
  • Constructed an executive Power BI dashboard, identifying 10 un-campaigned high-affinity households and evaluating an illustrative $18.1K sensitivity case.
PythonSQLPostgreSQLAWS S3AWS AthenaPower BI
GitHub Codebase
Risk Analytics & AIRisk Analytics

GST Invoice Anomaly Detection & Vendor Risk Scoring

Four-layer forensic framework (Detect → Score → Explain) with statistical anomaly detection and Gemini AI narratives in Tableau

An end-to-end invoice analytics and vendor risk solution analyzing 50K+ GST invoices across 210 vendors to detect anomalies, quantify vendor risk, and support audit prioritization.

Invoices Audited
50K+
High-Risk Vendors
32
AI Narratives
32
Technical Highlights
  • Engineered a four-layer risk architecture (Detect → Score → Explain) combining deterministic rule validation with Z-Score and IQR statistical anomaly detection.
  • Formulated deterministic vendor risk scoring across four signals (anomaly frequency, deviation magnitude, validation failures, recency), flagging 7,500+ invoices and 32 HIGH-risk vendors.
  • Integrated Gemini API to generate evidence-grounded vendor risk narratives from deterministic risk signals, without modifying risk scores or risk tiers.
  • Constructed an interactive drill-down Tableau forensic dashboard connecting 32 AI narratives with audit-recoverable tax capital.
PythonSQLPostgreSQLGemini APITableauRisk Scoring
Tableau DashboardGitHub Codebase
Data Quality & ReliabilityData Quality

E-Way Bill Data Reliability Pipeline

Automated reliability validation, KS and PSI statistical analysis, and Airflow PASS/FAIL gate for regulatory datasets

An analytical data-quality and reliability system evaluating 10,089 E-Way Bill records across FY2022-23 and FY2023-24 to ensure statistical integrity and data quality.

TrustedRecords
10,089
StatisticalResults
204
CorruptionScenarios
7
AutomatedTests
125
Technical Highlights
  • Built multi-layer reliability validation and cross-table reconciliation for 10,089 regulatory records across FY2022-23 and FY2023-24.
  • Conducted Kolmogorov-Smirnov (KS) tests and Population Stability Index (PSI) year-over-year statistical analysis generating 204 statistical results.
  • Designed an automated Airflow PASS/FAIL reliability gate blocking unvalidated data from promotion to trusted PostgreSQL storage.
  • Verified failure handling through 7 controlled corruption scenarios and 125 tests, blocking missing records, duplicates, invalid values, structural defects, and an INR 500 Cr alteration.
PythonSQLPostgreSQLApache AirflowDockerSciPyStreamlit
GitHub Codebase
Product Management · FintechPayment Reliability

Payment Status Recovery

UPI Transaction Uncertainty Engine

Designed a fintech prototype that resolves uncertain UPI payment outcomes using deterministic, evidence-based recovery logic. The system evaluates synthetic payment evidence across the remitter bank, payment switch, and beneficiary bank to distinguish confirmed failures from unresolved transactions and prevent unsafe duplicate retries.

Synthetic Cases
13
Banking Telemetry
4-Party
Automated Tests
88
Technical Highlights
  • Evidence-based transaction recovery with explicit retry-safety rules.
  • Transaction investigation and support-case management workflows.
  • Product analytics for recovery outcomes and support funnels.
Next.jsTypeScriptTailwind CSSVitest
Live DemoGitHub Codebase
Academic Foundation

Education

Quantitative grounding in mathematical analysis, systems thinking, and structured problem-solving.

B.Tech in Electronics and Communication Engineering

Birla Institute of Technology (BIT), Mesra
2022 – 2026Ranchi, Jharkhand, India

Developed a foundation in quantitative reasoning, mathematical analysis, and structured problem-solving, with experience applying analytical methods to complex technical problems.

Academic Foundations
Probability & Stochastic ProcessesMathematical Modeling & OptimizationStatistical AnalysisSystems ThinkingResearch & Experimental Validation
Quantitative Reasoning
Applying mathematical and statistical thinking to evaluate problems and interpret results.
Analytical Problem-Solving
Breaking complex problems into structured steps and evaluating possible solutions.
Research & Validation
Testing assumptions, comparing results, and investigating discrepancies to improve analytical reliability.
Technical Tooling

Skills & Capabilities

Focused toolchain built around analytics, statistical modeling, data workflows, and decision support.

Analytics

Statistical modeling, exploratory analysis, and quantitative querying

PythonSQLPandasNumPySciPyEDAStatistical Analysis

Risk & AI

Forensic anomaly identification, LLM APIs, and AI-assisted analytics

Risk AnalyticsAnomaly DetectionLLM APIsGenerative AIAPI IntegrationPrompt Engineering

Data Engineering & Cloud

Relational modeling, pipeline architecture, and cloud data querying

PostgreSQLETL/ELTData PipelinesData Quality & ValidationAWS S3AWS AthenaAirflow

BI & Tools

Interactive visual reporting, container orchestration, and version control

Power BITableauStreamlitDockerGitJupyter
DOMAIN ALIGNMENT

Areas of Interest

Focus areas spanning product execution, analytical problem-solving, risk evaluation, and operational improvement.

Product Strategy & ExecutionBusiness & Operational AnalysisRisk Analytics & Decision ScienceData-Driven Decision MakingAI-Assisted SolutionsProcess Improvement & Systems Reliability
Verified Qualifications & Honors

Credentials & Achievements

Selective industry credentials reinforcing cloud architectures, relational querying, and competitive engineering milestones.

Certifications & Credentials

Data & Cloud

AWS Certified Cloud Practitioner
2026
Amazon Web Services (AWS)

Cloud infrastructure, security, IAM, S3 storage architectures & billing management

Verified Qualification
AWS Fundamentals of Analytics (Part 1 & 2)
2026
Amazon Web Services (AWS)

Data lake design, distributed query optimization in Amazon Athena, ETL best practices

Verified Qualification

Analytics

Data Analytics Job Simulation
2026
Deloitte

Practical client risk data telemetry, forensic data cleaning, and executive reporting

Verified Qualification
SQL (Advanced) & Problem Solving
2026
HackerRank

Advanced joins, recursive queries, window functions, and relational indexing

Verified Qualification

Engineering & Electronics

AI & Deep Learning in Healthcare Workshop
2023
Institution of Electronics and Telecommunication Engineers (IETE)

Neural network architectures, biomedical signal classification, and feature engineering

Verified Qualification

Competitive Achievements

2023
Hackathon Finalist

Smart India Hackathon (SIH) 2023: 7th Position

Birla Institute of Technology (BIT), Mesra Internal Hackathon

Secured 7th position out of dozens of competitive engineering teams at the BIT Mesra internal round of Smart India Hackathon 2023, presenting a hardware-software integrated solution evaluated on technical feasibility, operational viability, and measurable impact.

Top 7 out of 40+ engineering teams at BIT Mesra
Get in Touch

Start a Conversation

Open to analytics, risk, product, operations, and data-focused opportunities. Reach out directly or send a message.

Need an Official Résumé?

To provide tailored versions aligned with specific role requirements, resumes are shared directly upon verified request.

Email Address
kumarsaksham560@gmail.com
Phone / WhatsApp
+91 9166552458
GitHub
LinkedIn

Direct Message

Send a note directly to my email regarding an opportunity or technical discussion.