M.S. Computer Science
University of Massachusetts Lowell • GPA 3.736 / 4 • Dec 2023
About
Hello! I’m Kamal — a Systems Engineer who enjoys taking ideas from whiteboard sketches to production‑ready solutions. My toolkit covers Python, Java, JavaScript, HTML/CSS, and SQL, and I’m most at home in a Linux terminal with Git close by.
Areas of focus
- Software Development: design, build, and maintain full‑stack applications.
- Machine Learning: image & text processing, deep & transfer learning.
- Database Management: schema design, performance tuning on MySQL, PostgreSQL, Redshift, Oracle.
- IoT with Raspberry Pi: prototype and deploy sensor‑driven lab solutions (novice‑expert level).
- REST‑centric Tooling: architect robust software that consumes, enriches, and surfaces data via APIs.
- Data Engineering: ETL with Informatica PowerCenter/IICS, SQL optimisation.
I like working where software, data, and hardware intersect — whether that’s automating lab workflows, wiring up IoT sensors, or shipping ML models behind clean REST endpoints. Day‑to‑day, you’ll find me iterating in VS Code or PyCharm, tracking work in JIRA, and keeping documentation tight in Markdown or Confluence.
Education
B.Tech Computer Science & Engineering
JNTU Hyderabad • GPA 6.73 / 10 • Sep 2020
Experience
Systems Engineer – Hooke Laboratories, LLC
Mar 2024 – Present- Led a team to digitise office operations with bespoke software, streamlining workflows and boosting efficiency.
- Designed & deployed machine‑learning models for tissue analysis, driving research precision and scalability.
- Built and maintained large‑language‑model pipelines that surface instant answers for Operations.
- Created instrument‑control software that optimises analytical workflows for research teams.
- Designed Raspberry‑Pi‑based IoT devices to monitor and secure lab equipment in real time.
- Automated project management with Google Apps Script, enhancing collaboration and data transparency.
- Served as company‑wide AI/ML adviser, evangelising best practices and guiding adoption.
- Strengthened IT operations & security by improving backup routines and incident response.
Facilities Information Systems Assistant – University of Massachusetts Lowell
Jan 2023 – Dec 2023- Engineered a Python codebase around e‑Builder REST API to extract and validate cost data, automating multi‑format reports.
- Refactored data import logic, cutting retrieval time by 70% and eliminating manual bottlenecks.
- Integrated heterogeneous data into MS SQL Server for reconciliation dashboards despite record‑limit constraints.
- Built HTML front‑ends and Tkinter dashboards to simplify work‑order management for non‑technical users.
- Developed ML‑powered pedestrian‑counting system with YOLO, reducing manual campus‑planning effort by 90%.
- Devised an intuitive interface for video uploads and region selection to enable self‑service analytics.
- Conducted unit & UAT testing, hardening the system’s stability and enhancing UX.
Programmer Analyst Trainee – Cognizant (Data Integration / ETL)
Dec 2020 – Nov 2021- Optimised data‑warehouse ETL for an automotive CRM (>5 M customers), improving query performance & processing speed.
- Designed complex mappings and workflows in Informatica PowerCenter, streamlining nightly loads.
- Administered PostgreSQL & Amazon Redshift back‑ends, reducing downtime by 10% through proactive maintenance.
- Authored comprehensive documentation to accelerate onboarding and facilitate troubleshooting.
- Collaborated with cross‑functional teams to ensure data accuracy and integrity across pipelines.
Skills
Programming Languages
- Python
- Java
- C/C++
- C#
- JavaScript
- HTML/CSS
- Google Apps Script
Databases
- Oracle SQL
- MySQL
- MariaDB
- MS SQL Server
- PostgreSQL
- Amazon Redshift
- PL/SQL
- T‑SQL
Python Libraries
- NumPy
- Pandas
- scikit‑learn
- TensorFlow
- Keras
- PyTorch
- OpenCV
- NLTK
- Matplotlib
Deployment Tools
- Flask
- Docker
- Streamlit
ML & AI Concepts
- Computer Vision
- NLP
- Deep Learning
- Transfer Learning
- Generative AI
- Prompt Engineering
ML Tools
- Hugging Face
- Haystack
- Ultralytics
ETL Tools
- Informatica PowerCenter
- Informatica IICS
Dev & Productivity
- Linux
- REST APIs
- Git
- JIRA
- PyCharm
- VS Code
- Jupyter
- Eclipse
- MS Office
- Google Workspace
Projects
Pedestrian Detection System using YOLO
- Spearheaded the development of a cutting-edge Pedestrian Detection System for the University of Massachusetts Lowell's Campus Planning Department.
- Created an intuitive UI to upload videos and draw detection regions; runs real‑time detect‑track‑count.
- Boosted planning efficiency by automating tallies that previously took hours.
Chest X‑Ray COVID‑19 Classifier
- Built deep‑learning pipeline (ResNet, VGG, LeNet) to train models capable of classifying chest X-ray images into four respiratory illnesses on 42k lung X-ray images.
- Implemented an advanced visualization technique GRADCAM to identify and highlight areas affected by the virus within the lungs, aiding medical professionals in targeted treatment approaches.
News Article Summarization: Evaluation of Cross-domain Adaptability of Text Summarizer
- Engineered advanced text summarizers, seamlessly integrating Extractive (TextRank) and Abstractive (BART) techniques.
- Enhanced BART Model performance significantly, achieving a 20% improvement over previous implementations.
- Streamlined a thorough evaluation of cross-domain adaptability, consistently outperforming benchmark ROUGE scores in contrast to the model's original implementation and a fine-tuned BBC News model.
Climate Change Sentiment Analysis
- Analysed 43k tweets on climate change; pipeline with NLTK preprocessing and Bi‑LSTM model.
- Employed Deep Learning Model with a Recurrent Neural Network approach using features including text in tweets and the frequency of specific keywords.
- Achieved 96% accuracy; validated robustness on fresh Twitter scrape.
Face Mask Detection
- CNN (Keras) model classifies Mask/No‑Mask with 90% accuracy on webcam streams.
- Real‑time alerts overlay & beep for non‑compliance using OpenCV.
Recolored Image Forgery Detection
- Implemented IEEE‑TIFS '19 deep discriminative model in TensorFlow to detect colour‑transfer forgeries.
- Tested on forged & internet images—accuracy improved 10% over baseline.
Citizens Income Prediction
- Benchmarked 6 ML algorithms on UCI Adult dataset (48k rows); Random Forest hit 92% accuracy.
- Delivered comparative metrics (precision +25%, recall +20% over baseline).
Diabetes Risk Prediction
- Trained XGBoost & Logistic models on Pima dataset; Streamlit app with SHAP explainability.
- Helps clinicians interpret top risk factors for preventive care.
Student Knowledge Modelling
- Predicts students’ mastery of DC Machines concepts; instance‑based classifier boosts F1 by 7%.
- GUI allows educators to upload CSVs and view predictions instantly.
Aadhaar‑Based Online Voting System
- Java Servlets/JSP + MySQL platform enabling secure remote voting; 20% turnout increase in pilot.
- Cut election costs 15% while preserving confidentiality & auditability.