01AI automation desktop app
AutoTube
YouTube automation software for AI-assisted scripts, video assembly, and publishing Shorts from one desktop workflow.
Akabir Abbas
I'm an AI and machine learning engineer building Python applications, LLM workflows, and AI automation. My software engineering work spans model monitoring, data pipelines, and desktop and web applications.
Open to AI, machine learning, and software engineering opportunities.
11 verified projects11 active projects · public and confidential work
Selected systems built for daily use.
Four projects that show practical engineering across operations, machine learning, and product software. Explore all projects.
02Assessment platform
Automated Evaluation System
A desktop-style exam workflow for OMR, essays, reports, and review.
03Reporting desktop app
RetailOps Reporting Suite
Turns raw POS exports into multi-sheet operational reports.
04ML monitoring platform
ModelOps Doctor
Model health, data drift, and diagnosis in one practical control room.
AI, ML & automation services
Explore what I can build for your team, with relevant project examples. View all services.
01
AI Application Development
Add LLM features to your software, connect AI APIs, and build prompt evaluation workflows so your team can review the output.
02
Machine Learning Development
Turn a dataset into a predictive model or computer vision tool, connect it to an application, and monitor how it performs.
03
Workflow Automation
Connect your business tools with Python, n8n, and REST APIs. Automate data processing, recurring reports, and steps your team repeats.
Projects I've built.
From automating YouTube Shorts to grading exam papers, here's a closer look at my work.
AutoTube
AI automation desktop appTurns a topic into a complete vertical video with AI-assisted scripting, media sourcing, narration, subtitles, FFmpeg rendering, YouTube upload, retries, and channel reporting.
Automated Evaluation System
Assessment platformCombines authentication, bubble-sheet processing, barcode detection, AI-assisted essay utilities, and report generation in one operational tool.
RetailOps Reporting Suite
Reporting desktop appReplaces manual spreadsheet work with a focused Windows workflow for store, employee, and protection-sales reporting.
ModelOps Doctor
ML monitoring platformRegisters models and datasets, ingests prediction logs, computes health scores, and schedules monitoring workflows with clear retraining guidance.
Shopkeeper POS
Offline operations systemTracks sales, purchases, inventory, credit, expenses, and profit & loss while keeping all data local to one machine.
DevMatch AI
Applied machine learningUses a KNN model, similar-developer matching, and tailored skill-gap guidance to make a large public dataset personally useful.
Red Bug Solution
Marketing websiteBuilt with the Next.js App Router and Tailwind, featuring a hand-written 2D particle engine rather than dependency-heavy WebGL.
PromptVault
Prompt operations APIOrganizes prompts and workspaces, runs evaluation jobs in the background, and combines durable storage with fast cached reads.
BigQuery Release Pulse
Release intelligenceParses and groups an XML feed into useful updates, with offline-aware caching, filtering, and a sharing composer.
Public Intelligence
Research prototypePairs a Next.js interface with a FastAPI, worker, and database stack; the project explicitly frames responsible, authorized use.
Smart Task Manager
Android productivity appA Kotlin and Jetpack Compose app with Room persistence, background reminder checks, and an operational dashboard.
Skills & Technologies
A practical toolkit spanning model development, product engineering, data infrastructure, and desktop delivery.
Machine learning
- Python
- scikit-learn
- Classification
- Regression
- Feature Engineering
- Model Training
- Model Evaluation
- Hyperparameter Tuning
- Predictive Modeling
Deep learning & computer vision
- PyTorch
- TensorFlow
- Keras
- Neural Networks
- CNNs
- Computer Vision
- OpenCV
- OCR
- OMR
- Image Processing
- Transfer Learning
- Model Optimization
- NLP
- Google Cloud Vision API
Generative AI & agents
- OpenAI API
- GPT-4.1
- Gemini API
- LangChain
- Hugging Face
- RAG Pipelines
- AI Agents
- Tool Calling
- Prompt Engineering
- Structured Outputs
- Embeddings
- Vector Search
Automation & integration
- n8n
- Workflow Automation
- Business Process Automation
- REST APIs
- Third-Party Integrations
- JSON Processing
- Celery
- Apache Airflow
- OAuth 2.0
Backend & data
- FastAPI
- Flask
- SQL
- Pandas
- NumPy
- ETL/ELT
- PostgreSQL
- MySQL
- MongoDB
- Redis
- SQLAlchemy
- Alembic
- SQLite
- Excel
- openpyxl
MLOps & delivery
- MLOps
- MLflow
- DVC
- Great Expectations
- Model Monitoring
- Drift Detection
- Docker
- GitHub Actions CI/CD
- Git
- pytest
- Linux/WSL2
- Structured Logging
- Kubernetes
Software & interfaces
- JavaScript
- TypeScript
- Java
- C/C++
- Kotlin
- Dart
- React
- Next.js
- Tailwind CSS
- Streamlit
- Electron
- CustomTkinter
- Eel
- PyInstaller
- FFmpeg
- Jetpack Compose
- Room
- WorkManager
How I work
From a real workflow to a dependable system.
I focus on the signal a person needs, then choose the smallest capable system that can deliver it.
Start with the real workflow.
Operations, reporting, assessment, and monitoring are designed around the actual decisions people need to make.
Choose the smallest capable stack.
From offline SQLite apps to queue-backed services, the architecture fits the environment rather than chasing novelty.
Make the important state visible.
Good interfaces surface priorities, drift, stock, deadlines, and report-ready outcomes before users need to ask.
Product systemsAI & machine learningData workflowsDesktop softwareWeb engineering
Let's connect
Build the next useful system.
Tell me what you're working on. I'll read your note and get back to you by email.
Project inquiry
Start a conversation.
Other ways to reach me