Open to opportunities

Sakshi Poudyal

@sakshipoudyal

Message

AI enthusiast skilled in machine learning and data visualization.

Nepal

Message

What I'm looking for

I am looking for opportunities that allow me to innovate in AI, collaborate with talented teams, and contribute to impactful projects.

I am a passionate AI and Machine Learning practitioner with hands-on experience in developing innovative models and solutions to tackle real-world problems. My proficiency in Python and frameworks such as PyTorch, TensorFlow, and Keras has enabled me to build robust applications that not only meet user needs but also push the boundaries of technology.

Throughout my academic and project work, I have honed my skills in data preprocessing, feature engineering, and model deployment. Notable projects include a neural steganography system that securely encrypts messages within images and a crime forecasting model that predicts crime hotspots using historical data. I am eager to contribute to cutting-edge AI research and development while continuously learning and growing in this dynamic field.

Experience

Work history, roles, and key accomplishments

Neural Steganography Developer

Self-Employed

Developed a neural system for encrypting and hiding sensitive messages within RGB images using an autoencoder architecture. Integrated AES-256 for secure pre-encryption and evaluated robustness against steganalysis models.

PyTorch TensorFlow Keras Python

Crime Forecasting ML Engineer

Self-Employed

Built an LSTM-based AI model to predict next-week crime hotspots from historical geospatial data, engineering data pipelines with GeoPandas and Shapely. Visualized predicted hotspots on interactive heatmaps using Folium and Matplotlib, deploying a Streamlit dashboard for real-time exploration.

Geopandas Shapely Matplotlib Streamlit Data Pipelines Geospatial Python

Autonomous Cipher Decryption AI Developer

Self-Employed

Designed and trained a hybrid 1D CNN + LSTM model for autonomous decryption of Caesar and Vigenère ciphers without prior key knowledge. Generated custom synthetic datasets and integrated language model verification using n-gram and RNN scoring for grammatical correctness.

1D CNN N Gram RNN Python

Education

Degrees, certifications, and relevant coursework

Nepal Engineering College

BSc in Computer Science and Engineering, Computer Science and Engineering

Studied core computer science concepts including data structures, algorithms, and programming technologies. Gained expertise in artificial intelligence, neural networks, image processing, and big data, alongside foundational knowledge in operating systems and computer networks.