Saurabh S.

Key Skills

Software

Other

Top Subject Matter

AI forensics and deepfake detection

KYC and face-swap/fraud detection

Multi-modal AI inference orchestration for media analysis

Top Data Types

Video

Image

Text

Document

Top Task Types

Object Detection

Bounding Box

Segmentation

Data Collection

Prompt + Response Writing (SFT)

Computer Programming/Coding

Freelancer Overview

PaladinAi — Software Engineer (AI forensics/deepfake detection and transcription systems). Brings 3+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Other. Education includes Bachelor of Technology, ABESIT (2024) and Class XII, St. Paul's Senior Secondary School (2022). AI-training focus includes data types such as Video and labeling workflows including Object Detection, Evaluation, and Rating.

Labeling Experience

PaladinAi — Software Engineer (AI forensics/deepfake detection and transcription systems)

OtherVideoObject Detection

Worked on an AI-based forensics and deepfake detection system, focusing on media analysis pipelines. Designed scalable backend services and data processing workflows to support real-time inference on sensitive forensic video data. Optimized inference throughput using GPU acceleration and event-driven task prioritization mechanisms. • Implemented audio/video detection components for transcription and translation phonetic systems. • Introduced Kafka and Flink-style architecture for task queue prioritization. • Ensured secure handling and processing of sensitive forensic inputs. • Built scalable database-backed services for AI inference workloads.

2025 - Present

High-Performance AI Inference Orchestration Engine — Personal Project (Rust)

OtherVideo

Engineered a high-concurrency multi-modal AI inference orchestration engine for deepfake detection and media analysis. Implemented device-selection logic to optimize inference across CUDA (NVIDIA), MPS (Apple Silicon), and CPU environments. Managed task lifecycles and progress reporting to support long-running inference workflows. • Built asynchronous REST APIs for large multi-part file uploads and deep learning task orchestration. • Integrated Torch (tch-rs) and OpenCV to process video frames and audio streams. • Implemented background workers with heartbeat monitoring and zombie-task recovery. • Added frame-accurate progress reporting and typed OpenAPI/Swagger model discovery.

2024 - 2024

PaladinAi — Backend Developer (real-time KYC + deepfake/face-swap detection)

OtherVideoObject Detection

Built a real-time KYC and fraud detection pipeline involving frame extraction and face-swap detection. Optimized backend pipelines for high-speed inference and accurate fraud detection at scale. Designed and deployed microservice-based architecture and database systems tailored for low-latency AI workloads. • Utilized FFmpeg for efficient frame extraction enabling real-time processing. • Implemented database schemas, indexing, and async SQL queries via SQLx. • Developed high-performance Rust services using Tokio and SQLx for type-safe data access. • Supported scalable, secure inference operations for continuous detection tasks.

2024 - 2024

Education

Work History