diff --git a/.gitignore b/.gitignore
index 1ef1347..4c02b47 100644
--- a/.gitignore
+++ b/.gitignore
@@ -27,4 +27,31 @@ __pycache__/
 
 # OS
 .DS_Store
-Thumbs.db
\ No newline at end of file
+Thumbs.db
+
+# Outputs
+outputs/
+
+# Models
+*.pth
+*.pt
+*.ckpt
+
+# Logs
+*.log
+
+# Cache
+__pycache__/
+*.pyc
+
+# VSCode
+.vscode/
+
+# Conda
+.env
+.venv
+
+# Dataset
+data/raw/
+data/intermediate/
+data/processed/
\ No newline at end of file
diff --git a/PROJECT_STATUS.md b/PROJECT_STATUS.md
new file mode 100644
index 0000000..e5797bd
--- /dev/null
+++ b/PROJECT_STATUS.md
@@ -0,0 +1,68 @@
+# DeepVision AI v2.0
+
+## Sprint 1 – Foundation ✅
+
+- [x] Repository Recovery
+- [x] GitHub Repository
+- [x] Development Environment
+- [x] Configuration System
+
+---
+
+## Sprint 2 – Data Pipeline ✅
+
+- [x] Dataset Index Builder
+- [x] Dataset CSV Generator
+- [x] Sequence Dataset
+
+---
+
+## Sprint 3 – Training Infrastructure
+
+- [ ] Balanced Sampler
+- [ ] DataLoader
+- [ ] Trainer
+- [ ] Checkpoint Manager
+- [ ] Logger
+
+---
+
+## Sprint 4 – Model
+
+- [ ] ViT Integration
+- [ ] EfficientNet-BiLSTM
+- [ ] Model Factory
+
+---
+
+## Sprint 5 – Evaluation
+
+- [ ] Metrics
+- [ ] ROC Curve
+- [ ] Confusion Matrix
+- [ ] Cross Dataset Evaluation
+
+---
+
+## Sprint 6 – Explainability
+
+- [ ] Grad-CAM
+- [ ] Heatmap Overlay
+- [ ] Localization Metrics
+
+---
+
+## Sprint 7 – Deployment
+
+- [ ] Streamlit App
+- [ ] REST API
+- [ ] Docker
+
+---
+
+## Sprint 8 – Documentation
+
+- [ ] README
+- [ ] Architecture Diagram
+- [ ] Research Report
+- [ ] Portfolio Assets
\ No newline at end of file
diff --git a/README.md b/README.md
index e69de29..62a1597 100644
--- a/README.md
+++ b/README.md
@@ -0,0 +1,20 @@
+# DeepVision AI
+
+An Explainable Deepfake Detection Framework using Vision Transformers, Temporal Modeling, and Grad-CAM.
+
+## Status
+
+🚧 Under Active Development
+
+## Features
+
+- Vision Transformer based Deepfake Detection
+- Explainable AI using Grad-CAM
+- FF++ and Celeb-DF Support
+- Cross Dataset Evaluation
+- Streamlit Deployment
+- Research-oriented Modular Framework
+
+## Repository Structure
+
+See PROJECT_STATUS.md for the current development roadmap.
\ No newline at end of file
diff --git a/configs/default.yaml b/configs/default.yaml
new file mode 100644
index 0000000..5ffa1f9
--- /dev/null
+++ b/configs/default.yaml
@@ -0,0 +1,42 @@
+project:
+  name: DeepVision AI
+  version: 2.0
+
+paths:
+  raw_data: data/raw
+  processed_data: data/processed
+  sequences: data/intermediate/sequences
+  checkpoints: outputs/checkpoints
+  logs: outputs/logs
+  metrics: outputs/metrics
+  predictions: outputs/predictions
+
+dataset:
+  name: FFPP_CelebDF
+  image_size: 224
+  sequence_length: 30
+  fps: 5
+  batch_size: 8
+  num_workers: 4
+
+training:
+  epochs: 30
+  learning_rate: 0.0001
+  weight_decay: 0.0001
+  optimizer: AdamW
+  scheduler: CosineAnnealingLR
+
+model:
+  backbone: vit_b_16
+  pretrained: true
+  freeze_backbone: true
+  temporal_pooling: mean
+  num_classes: 2
+
+evaluation:
+  metrics:
+    - accuracy
+    - precision
+    - recall
+    - f1
+    - auc
\ No newline at end of file
diff --git a/docs/ARCHITECTURE.md b/docs/ARCHITECTURE.md
new file mode 100644
index 0000000..e9e342a
--- /dev/null
+++ b/docs/ARCHITECTURE.md
@@ -0,0 +1,447 @@
+# DeepVision AI v2.0 Architecture
+
+> A Modular Explainable Deepfake Detection Framework using Vision Transformers, Temporal Modeling, and Explainable AI.
+
+---
+
+# Overview
+
+DeepVision AI is a research-oriented and production-ready framework for detecting manipulated facial videos using deep learning.
+
+The framework is designed around modular engineering principles, allowing datasets, models, training pipelines, explainability techniques, and deployment components to evolve independently.
+
+---
+
+# System Objectives
+
+The primary objectives of DeepVision AI are:
+
+- Detect AI-generated deepfake videos.
+- Support multiple benchmark datasets.
+- Provide explainable predictions using Grad-CAM.
+- Enable reproducible deep learning experiments.
+- Provide a deployable real-time inference system.
+
+---
+
+# System Architecture
+
+```
+                        DeepVision AI
+
+                    Raw Video Dataset
+                           │
+                           ▼
+                Dataset Index Generator
+                           │
+                           ▼
+               outputs/dataset_index.csv
+                           │
+                           ▼
+                 Sequence Dataset Loader
+                           │
+                           ▼
+                 Balanced Data Sampler
+                           │
+                           ▼
+                     PyTorch DataLoader
+                           │
+                           ▼
+                     Training Engine
+                           │
+           ┌───────────────┴───────────────┐
+           ▼                               ▼
+ EfficientNet + BiLSTM          Vision Transformer
+           │                               │
+           └───────────────┬───────────────┘
+                           ▼
+                    Classification Head
+                           │
+                           ▼
+                     Model Checkpoints
+                           │
+          ┌────────────────┴────────────────┐
+          ▼                                 ▼
+     Evaluation Engine                Inference Engine
+          │                                 │
+          ▼                                 ▼
+      Performance                     Grad-CAM
+      Metrics                         Visualization
+          │                                 │
+          └────────────────┬────────────────┘
+                           ▼
+                    Streamlit Application
+```
+
+---
+
+# Repository Structure
+
+```
+DeepVision-AI/
+
+├── app/
+│
+├── assets/
+│
+├── configs/
+│
+├── data/
+│   ├── raw/
+│   ├── intermediate/
+│   └── processed/
+│
+├── docs/
+│
+├── outputs/
+│
+├── src/
+│
+├── tests/
+│
+├── train.py
+├── evaluate.py
+├── predict.py
+└── README.md
+```
+
+---
+
+# Source Code Architecture
+
+```
+src/
+
+datasets/
+    dataset_indexer.py
+    sequence_dataset.py
+    balanced_sampler.py
+    dataloader.py
+
+models/
+    vit_temporal_pooling.py
+    efficientnet_bilstm.py
+    model_factory.py
+
+training/
+    trainer.py
+    checkpoint.py
+    early_stopping.py
+
+evaluation/
+    evaluator.py
+    metrics.py
+
+explainability/
+    gradcam.py
+
+inference/
+    predictor.py
+
+utils/
+    config.py
+    logger.py
+```
+
+---
+
+# Data Pipeline
+
+```
+Raw Videos
+      │
+      ▼
+Video Indexing
+      │
+      ▼
+Sequence Generation (.npy)
+      │
+      ▼
+Sequence Dataset
+      │
+      ▼
+Balanced Sampling
+      │
+      ▼
+PyTorch DataLoader
+      │
+      ▼
+Model Training
+```
+
+---
+
+# Model Pipeline
+
+```
+Sequence
+
+↓
+
+Vision Transformer
+
+↓
+
+Temporal Pooling
+
+↓
+
+Classification Head
+
+↓
+
+Prediction
+```
+
+---
+
+# Training Pipeline
+
+```
+Load Configuration
+
+↓
+
+Load Dataset
+
+↓
+
+Create DataLoader
+
+↓
+
+Initialize Model
+
+↓
+
+Training Loop
+
+↓
+
+Validation
+
+↓
+
+Checkpoint Saving
+
+↓
+
+Performance Metrics
+```
+
+---
+
+# Evaluation Pipeline
+
+```
+Model
+
+↓
+
+Test Dataset
+
+↓
+
+Predictions
+
+↓
+
+Accuracy
+
+↓
+
+Precision
+
+↓
+
+Recall
+
+↓
+
+F1 Score
+
+↓
+
+ROC Curve
+
+↓
+
+Confusion Matrix
+```
+
+---
+
+# Explainability Pipeline
+
+```
+Input Sequence
+
+↓
+
+Forward Pass
+
+↓
+
+Target Layer
+
+↓
+
+Grad-CAM
+
+↓
+
+Heatmap
+
+↓
+
+Overlay
+
+↓
+
+Visualization
+```
+
+---
+
+# Deployment Pipeline
+
+```
+Upload Video
+
+↓
+
+Face Extraction
+
+↓
+
+Sequence Generation
+
+↓
+
+Model Prediction
+
+↓
+
+Grad-CAM
+
+↓
+
+Confidence Score
+
+↓
+
+Download Report
+```
+
+---
+
+# Engineering Principles
+
+The project follows the following principles:
+
+- Modular Architecture
+- Configuration Driven
+- Test Driven
+- Reproducible Experiments
+- Clean Separation of Responsibilities
+- Research Friendly
+- Deployment Ready
+
+---
+
+# Development Workflow
+
+```
+feature/*
+
+↓
+
+develop
+
+↓
+
+main
+```
+
+Each feature is implemented independently, tested, reviewed, and merged into the development branch before being promoted to the main branch.
+
+---
+
+# Supported Datasets
+
+Current datasets:
+
+- FaceForensics++
+- Celeb-DF v2
+
+Planned support:
+
+- DFDC
+- WildDeepfake
+- ForgeryNet
+- DeeperForensics-1.0
+
+---
+
+# Current Progress
+
+## Sprint 1
+
+- Repository Recovery
+- Development Environment
+- Configuration System
+
+## Sprint 2
+
+- Dataset Index Builder
+- Sequence Dataset
+- Dataset Metadata Generator
+
+## Sprint 3
+
+- Balanced Sampler (In Progress)
+- DataLoader
+- Training Engine
+
+## Sprint 4
+
+- Vision Transformer Integration
+- EfficientNet-BiLSTM
+- Model Factory
+
+## Sprint 5
+
+- Evaluation Pipeline
+
+## Sprint 6
+
+- Explainability
+
+## Sprint 7
+
+- Streamlit Deployment
+
+---
+
+# Future Improvements
+
+- Multi-GPU Training
+- Mixed Precision Training
+- Distributed Training
+- Self-Supervised Pretraining
+- Transformer Ensemble
+- Model Quantization
+- ONNX Export
+- TensorRT Deployment
+
+---
+
+# Citation
+
+If you use this framework in research, please cite the repository after publication.
+
+---
+
+# License
+
+This project will be released under the MIT License.
\ No newline at end of file
diff --git a/environment.yml b/environment.yml
index 3d86ea6..7db3e6d 100644
--- a/environment.yml
+++ b/environment.yml
@@ -1,4 +1,22 @@
-#python -m src.preprocessing.extract_frames --datasets ffpp
-#python -m src.preprocessing.extract_frames --datasets celebdf
-#python -m src.preprocessing.face_align_mtcnn
-#python -m src.training.train_frame 
\ No newline at end of file
+name: deepvision-ai
+
+channels:
+  - pytorch
+  - conda-forge
+
+dependencies:
+  - python=3.10
+  - pip
+
+  - numpy
+  - pandas
+  - matplotlib
+  - scikit-learn
+  - scikit-image
+  - pyyaml
+  - tqdm
+
+  - pip:
+      - opencv-python
+      - facenet-pytorch
+      - timm
\ No newline at end of file
diff --git a/src/__init__.py b/src/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/evaluation/__init__.py b/src/evaluation/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/explainability/__init__.py b/src/explainability/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/inference/__init__.py b/src/inference/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/models/__init__.py b/src/models/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/preprocessing/__init__.py b/src/preprocessing/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/training/__init__.py b/src/training/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/utils/__init__.py b/src/utils/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/utils/config.py b/src/utils/config.py
new file mode 100644
index 0000000..536a32d
--- /dev/null
+++ b/src/utils/config.py
@@ -0,0 +1,63 @@
+"""
+Configuration utilities for DeepVision AI
+
+Loads and validates YAML configuration files.
+"""
+
+from pathlib import Path
+import yaml
+
+
+def load_config(config_path: str = "configs/default.yaml"):
+    """
+    Load YAML configuration file.
+
+    Args:
+        config_path (str): Path to YAML configuration.
+
+    Returns:
+        dict: Configuration dictionary.
+    """
+
+    config_path = Path(config_path)
+
+    if not config_path.exists():
+        raise FileNotFoundError(
+            f"Configuration file not found: {config_path}"
+        )
+
+    with open(config_path, "r", encoding="utf-8") as file:
+        config = yaml.safe_load(file)
+
+    return config
+
+
+def get_path(config, key):
+    """
+    Return project path from config.
+    """
+
+    return Path(config["paths"][key])
+
+
+def print_config(config):
+    """
+    Nicely print configuration.
+    """
+
+    print("=" * 60)
+    print(" DeepVision AI Configuration")
+    print("=" * 60)
+
+    for section, values in config.items():
+
+        print(f"\n[{section}]")
+
+        if isinstance(values, dict):
+            for k, v in values.items():
+                print(f"{k:20} : {v}")
+
+        else:
+            print(values)
+
+    print("=" * 60)
\ No newline at end of file
diff --git a/tests/test_config.py b/tests/test_config.py
new file mode 100644
index 0000000..27b4b4a
--- /dev/null
+++ b/tests/test_config.py
@@ -0,0 +1,6 @@
+from src.utils.config import load_config, print_config
+
+
+config = load_config()
+
+print_config(config)
\ No newline at end of file
diff --git a/tests/test_dataset_builder.py b/tests/test_dataset_builder.py
new file mode 100644
index 0000000..befa853
--- /dev/null
+++ b/tests/test_dataset_builder.py
@@ -0,0 +1,15 @@
+from src.datasets.tf_dataset_builder import DatasetIndexer
+
+builder = DatasetIndexer("data/raw")
+
+samples = builder.scan()
+
+builder.summary()
+
+# Save dataset index
+builder.save_index("outputs/dataset_index.csv")
+
+print("\nFirst 10 Samples\n")
+
+for sample in samples[:10]:
+    print(sample)
\ No newline at end of file
diff --git a/tests/test_sequence_dataset.py b/tests/test_sequence_dataset.py
new file mode 100644
index 0000000..f14b9af
--- /dev/null
+++ b/tests/test_sequence_dataset.py
@@ -0,0 +1,24 @@
+from src.datasets.sequence_dataset import SequenceDataset
+
+dataset = SequenceDataset(
+    csv_file="outputs/dataset_index.csv",
+    sequence_root="data/intermediate/sequences"
+)
+
+print("Dataset Size:", len(dataset))
+
+sample = dataset[0]
+
+print()
+
+print("Sequence Shape :", sample["sequence"].shape)
+print("Sequence Type  :", sample["sequence"].dtype)
+print("Label          :", sample["label"])
+print("Dataset        :", sample["dataset"])
+print("Video          :", sample["video_name"])
+
+print()
+
+print("Min:", sample["sequence"].min())
+
+print("Max:", sample["sequence"].max())
\ No newline at end of file