AI Engineer - Agentic Video Generation

Company : TrueFan AI

Location : Gurgaon, India (On-site)

About TrueFan AI :

TrueFan AI is a generative-AI platform building AI-powered celebrity video ads and hyper-personalized marketing solutions for Indian brands. We work at the intersection of GenAI, video, voice, and automation, building production systems that create content at scale.

About the Role :

We are looking for an AI Engineer experienced in building large-scale, agent-driven video generation pipelines. You will build systems where AI agents orchestrate multiple models and tools to take a video brief from idea - script - scenes - generation - voice/lip-sync - editing - validation - final video.

What You'll Do :

Agentic Video Generation :

- Build multi-step AI agents that autonomously plan and execute video-generation workflows.

- Orchestrate LLMs, image/video generation models, TTS, voice, lip-sync, avatars, and other AI tools.

- Design agent state, tool calling, planning, validation, retries, fallbacks, and human-in-the-loop workflows.

- Build systems that can dynamically decide the next step based on intermediate outputs.

Scale & Infrastructure :

- Build high-throughput video-generation pipelines capable of processing large batches and concurrent jobs.

- Design asynchronous workflows using queues, workers, APIs, and distributed systems.

- Optimize pipelines for quality, latency, throughput, reliability, and cost.

- Build robust handling for failures, retries, timeouts, partial outputs, and long-running jobs.

AI & Video Engineering :

- Integrate and evaluate new generative video, image, audio, and multimodal models.

- Build automated video-processing workflows using tools such as FFmpeg.

- Develop pipelines for compositing, rendering, subtitles, audio/video synchronization, and asset management.

- Build evaluation and quality-control systems for generated content.

Production Engineering :

- Build production-grade Python services and APIs.

- Implement monitoring, logging, tracing, error handling, and automated validation.

- Debug issues across the entire AI pipeline, from agent decisions to model failures and rendering jobs.

- Take systems from prototype - production - scale.

What We're Looking For :

- 2 - 4 years of experience in AI/ML, software engineering, or generative AI.

- Hands-on experience building video-generation or generative-media pipelines.

- Strong Python and backend engineering fundamentals.

- Experience building LLM-powered agents and multi-step workflows.

- Experience integrating multiple AI models/APIs into production pipelines.

- Strong understanding of asynchronous processing, queues, workers, and distributed systems.

- Ability to build reliable systems at scale with a focus on latency, cost, and throughput.

- Strong debugging, problem-solving, and ownership mindset.

Nice to Have :

- Experience with LangGraph, LangChain, CrewAI, or similar agent frameworks.

- Experience with video-generation models, TTS, voice cloning, lip-sync, avatars, or multimodal AI.

- Experience with FFmpeg, GPU inference, or distributed GPU workloads.

- Experience with Celery, RabbitMQ, Redis, Kafka, Docker, or cloud infrastructure.

- Experience building evaluation systems for generative AI.

Our Stack :

- AI / Agents : LLM APIs, Agent Frameworks, Multimodal AI

- Generative Media : Video, Image, Voice, TTS, Lip-sync

- Backend : Python, FastAPI

- Video Processing : FFmpeg

- Orchestration : Celery, RabbitMQ, Redis, Background Workers

- Infrastructure : Docker, Cloud / GPU Workloads

- Database : MySQL / PostgreSQL

What Success Looks Like :

- Build and ship agentic video-generation pipelines at scale.

- Automate the journey from a video brief to a finished video with minimal manual intervention.

- Reliably orchestrate multiple AI models and tools.

- Improve video quality, pipeline speed, reliability, and cost.

- Build infrastructure capable of handling large volumes of concurrent video-generation jobs.

- Rapidly integrate and productionize new AI models and capabilities.