ANUJ
DATA×AI×ENGINEERING
INITIALIZING SYSTEM
ANUJ MUNDU
RETURN TO PORTFOLIO

ANUJ MUNDU

PYTHON BACKEND ENGINEER · AI/ML SYSTEMS · DATA SCIENTIST
anujmark.edwin.ame@gmail.com•Bhopal, India (Open to Remote / Relocation)•github.com/anujmundu•linkedin.com/in/anujmundu

PROFESSIONAL SUMMARY

Applied AI/ML Systems Engineer, Data Scientist, and Python Backend Architect specializing in the end-to-end lifecycle: from SQL database extraction and exploratory data analysis to statistical predictive modeling in Scikit-Learn, deep learning computer vision in PyTorch, and low-latency microservice deployment via FastAPI and Docker.

CORE TECHNICAL SKILLS

01. Data Analytics: SQL (PostgreSQL), DuckDB Columnar OLAP, Python, Pandas, NumPy, Statistics, EDA, Data Cleaning, Star Schema, Cohort Retention.
02. Data Science & MLOps: Scikit-Learn, XGBoost, Feature Engineering, Calibrated Cross-Validation, SHAP Factor Attribution, ROC-AUC, Purged K-Fold.
03. AI & Computer Vision: PyTorch 2.5, OpenCV, YOLOv5-CASP, Torchvision, Transfer Learning, ResNet-50, Grad-CAM, INT8 ONNX Runtime Quantization.
04. Distributed Backend & Deploy: FastAPI, Redis Task Queues, RabbitMQ, Multi-Stage Docker, Hexagonal Architecture, Git, Linux CLI, CI/CD.

FLAGSHIP ENGINEERING PROJECTS

PulseMetrics Copilot™ — SaaS Revenue & Cohort IntelligenceQuery Speed: < 1.2s
Technologies: DuckDB, Python 3.11+, Streamlit, Plotly, Scikit-Learn, SQLite3 Fallback, Text-to-SQL, Cohort Analysis

Enterprise SaaS revenue intelligence platform engineered with an embedded DuckDB columnar engine that aggregates 541k+ rows in <1.2s. Implements dynamic cohort retention matrices (M0 to M12+), MRR waterfall decompositions, flight-risk churn scoring, and natural language Text-to-SQL.

OmniVision DocIntel API™ — Document Forensics & ELAP95 Latency: < 65ms
Technologies: Python 3.11+, FastAPI Async, OpenCV, Error Level Analysis, Pydantic v2, Docker, Prometheus, Streamlit, SSE

Production-grade asynchronous FastAPI microservice engineered to defend against digital document fraud (altered numbers, spliced receipts, forged credentials). Implements OpenCV camera blur detection, orientation skew correction, Error Level Analysis (ELA forensics), and token-bucket rate limiting in a sub-200MB Docker container.

Multi-Paradigm Vision Benchmark & ONNX ArenaTop-5 Accuracy: 98.92%
Technologies: PyTorch 2.6+, CUDA 12 AMP, ONNX Runtime INT8, ResNet, ConvNeXt, EfficientNet, Vision Transformer, Swin Transformer, FastAPI, Streamlit

Production-grade deep learning vision benchmark evaluating 11 distinct neural network architectures across 5 foundational inductive bias paradigms on 50 object categories. Features Soft-Voting Ensemble (98.92% Top-5), ONNX INT8 Quantization (3.01x speedup), Grad-CAM explainability, and a live Streamlit Arena.

Autonomous AI CFO & Profit Intelligence SuiteML Tournament: 10 Algorithms
Technologies: Python 3.11+, Streamlit, FastAPI, Scikit-Learn, SHAP, SciPy SLSQP, Monte Carlo, Financial Modeling, Docker

Enterprise-grade financial intelligence, multi-model predictive analytics, and prescriptive capital allocation platform. Features a 10-algorithm machine learning tournament, SHAP explainability, 10,000-iteration Monte Carlo risk simulation, SLSQP inverse goal-seeking, liquidity stress testing, and an autonomous 3-agent AI CFO boardroom.

AutoRecon Agentic™ — Financial Reconciliation & AuditHours Saved: 25+ hrs/wk
Technologies: Python 3.11+, Streamlit, Pandas, PyPDF, OpenPyXL, TheFuzz, Rich, Pydantic v2, FinTech Automation

Production-grade autonomous financial reconciliation and dispute automation platform. Eliminates 25+ weekly hours of manual accounts payable matching by ingesting multi-format invoices (PDF, Excel, CSV), applying Levenshtein token-sort fuzzy matching, and autonomously drafting legally grounded dispute notices across 3 professional tones.

OmniForge — Production Multimodal AI PlatformP95 API Latency: 38.4ms
Technologies: Python 3.11+, FastAPI, PyTorch, Celery, Redis Queue, Kubernetes, Helm, Docker, Agentic RAG, Streamlit

Enterprise-grade multimodal intelligence platform engineered with FastAPI, PyTorch, Celery, Redis queue, and Kubernetes Helm deployments. Unifies Classical ML, Deep Vision, NLP, RAG, and automated adversarial prompt defense.

YOLOv5-CASP Clinical CADx — Lung Nodule SuiteSensitivity: 94.2%
Technologies: PyTorch 2.5.1, CUDA 12.1, OpenCV 4.9.0, YOLOv5-CASP, CBAM, ASPP, CoT3, DICOM, PACS, Streamlit

Clinical AI Computer-Aided Diagnosis (CADx) suite for pulmonary nodule detection in Chest X-Ray and CT scans. Enhanced YOLOv5 architecture integrating CBAM attention, ASPP multi-scale context, and CoT3 contextual transformers with Lung-RADS PACS workstation.

EXPERIENCE & EDUCATION

Python Backend & AI/ML Systems Engineer2023 — Present
Open-Source AI Engineering & Systems Development — Bhopal, India
  • Architected production multimodal platforms (OmniForge) integrating Agentic RAG, neural vision, Celery-Redis distributed queues, and Kubernetes Helm deployments.
  • Engineered clinical CADx suites (YOLOv5-CASP) with CBAM attention, ASPP context, and CoT3 transformers for automated pulmonary nodule localization.
  • Built production workforce MLOps platforms (RetainAI) using FastAPI, PyTorch Tabular ResNet/VAE, TreeSHAP explainability, and Kolmogorov-Smirnov statistical drift detection.
  • Developed high-reliability distributed task engines with decoupled worker daemons, Redis queues, and durable PostgreSQL state machines.
Master of Computer Applications (MCA)2023 — 2026
Maulana Azad National Institute of Technology (MANIT) — Bhopal, Madhya Pradesh, India
  • Premier National Institute of Technology (NIT) curriculum covering Advanced Data Structures & Algorithms, Distributed Systems, Cloud Computing, Database Management Systems, Machine Learning, and Computer Vision.
  • Authored research and engineering projects on Automated Multi-Modal Lung Nodule Detection using YOLOv5-CASP with attention mechanisms and real-time medical imaging workstations.
  • Engineered end-to-end full-stack AI/ML systems including distributed worker queues, agentic reasoning microservices, and statistical diagnostic pipelines.
Bachelor of Science (Honours) in Computer Science2020 — 2023
Guru Ghasidas Vishwavidyalaya (Central University) — Bilaspur, Chhattisgarh, India
  • Graduated with First-Class Honours with core focus on Computer Science fundamentals, Object-Oriented Programming, Discrete Mathematics, and Relational Databases.
  • Developed foundational algorithmic problem-solving capabilities, database schema design, and Linux systems administration.
  • Implemented statistical data analysis, scientific computation, and machine learning models for academic research projects.