Skills
Languages
Python · SQL · Bash / Linux · R · C++
Ranking & RecSys
Multi-task learning (CTR/CVR/CTCVR), Deep & Cross Networks (DCN-v2), Feature-Tokenizer Transformers, Mixture-of-Experts, candidate generation / two-tower models, learning-to-rank, propensity modelling
Computer Vision
Semantic segmentation (DeepLabV3+, U-Net, Attention U-Net), classification (ResNet, EfficientNet, MobileNetV3), attention mechanisms, augmentation strategies
Generative AI & NLP
LLM pipelines, prompt engineering, RAG, fine-tuning, embeddings, information extraction, knowledge-graph construction — spaCy, HuggingFace Transformers, LangChain
Evaluation & Explainability
NDCG, AUC-ROC, hit-rate, precision/recall, IoU/Dice, A/B testing, offline↔online correlation, SHAP, LIME
ML Engineering & Infra
PyTorch, TensorFlow, scikit-learn, ONNX, Triton Inference Server, distributed training (TorchDistributor, 8-GPU), Databricks (PySpark, Delta Lake, MLflow), Airflow, feature stores
Cloud & Data
Google Cloud (GCS, Vertex AI), Microsoft Azure, Docker, Kubernetes, PySpark, pandas, NumPy, SQL, Delta tables, MongoDB, Qdrant
Bioinformatics
QTLtools, BCFtools, eQTL mapping, statistical genetics, genomic data QC, biomedical literature mining, knowledge graphs, NetworkX, CytoscapeJS
Visualisation & Apps
Matplotlib · Seaborn · Plotly · Streamlit · Flask · React · Bootstrap · HTML/CSS · Weights & Biases · Git/GitHub · CI/CD (GitHub Actions)