w
wasilbhatti2

Wasil Bhatti

@wasilbhatti2

AI Engineer

Paquistão
Inglês, Urdu
Algumas informações são exibidas no idioma inglês.
Sobre mim
I build AI systems that work on real data. What I can do for you: - Computer vision: object detection, tracking, number plate reading, safety/PPE checks, image processing - Document AI: OCR for invoices, receipts and bills - AI agents: voice agents, chatbots, RAG, n8n automations - Machine Learning: price prediction, sales forecasting, churn, fault and anomaly detection - Data Analytics: Power BI dashboards, SQL, Python - Web and Mobile apps around these I check your data first and tell you honestly what accuracy is realistic. You get the trained model and full source code.... Saiba mais

Habilidades

w
wasilbhatti2
Wasil Bhatti
offline • 
Tempo médio de resposta: 1 hora

Conheça meus serviços

Integrações de IA
I will build a custom computer vision model for object detection and tracking
A partir deUS$ 90 /projetoMais detalhes
Integrações de IA
I will build an ai voice agent that answers calls and books appointments
A partir deUS$ 90 /projetoMais detalhes

Portfólio

Experiência profissional

Co-founder & AI Engineer

TwinStack Studio • Período integral

Jul 2026 - Present • 3 mos

Co-founded a small studio building web, mobile and AI systems for clients. I lead the AI side: computer vision (YOLO object detection, tracking, OCR), machine learning models (XGBoost, scikit-learn, PyTorch) and data analysis. Built a number plate recognition system with our own OCR model, an ML fault-detection system for air-handling units, an ML stock-ranking engine with backtesting, and an image-processing app, all deployed live with FastAPI.

VisionRD

Computer Vision Engineer

VisionRD • Período integral

Jun 2025 - Jul 2026 • 1 yr 1 mo

• Developed an image generation API using Stable Diffusion and other diffusion models, with GPU memory optimization • Researched and implemented multi-GPU model inference to reduce latency and memory footprint • Translated cutting-edge research into production by implementing layer caching techniques from recent papers, accelerating inference pipelines for advanced deep learning models. • Developed a multi-head deep learning model for automated PCB component inspection (98.2% accuracy) • Fine-tuned OCR models for PCB value recognition, significantly improving font-specific accuracy and enabling more reliable automated inspections. • Explored video-based generative and vision models for automotive inspection tasks, extending AI capabilities from static image analysis to dynamic, real-time inspection.

Central_Power Purchasing Agency

Data Analyst

Central Power Purchasing Agency • Período integral

Apr 2024 - May 2025 • 1 yr 1 mo

• Automated data cleaning and ETL pipelines in Python, reducing manual preparation time by 70% and ensuring high-quality, standardized datasets. • Developed time-series forecasting models for national power consumption, delivering more accurate predictions that informed critical energy-sector decision-making. • Built a data upload bot to streamline portal integration, improving data management efficiency and supporting organization-wide reporting.