About The Client The client is a new AI startup founded by people who’ve spent decades in the broadcast and media world. They are building an AI‑driven video platform that can automatically generate highlights, promos, and story‑driven clips - something the industry has needed for years but never had the technology to do well.

They're early. No website yet. No corporate layers. Just a small, committed team building from scratch with deep industry backing and a very real problem to solve.

If you want to learn fast, build fast, and be part of the founding tech team, this is the place.

Role: Junior AI Engineer As a Junior AI Engineer , you’ll work directly with the Senior AI Engineer and AI Architect to prototype, experiment, and help build the core AI engine.

This is a hands-on role where you’ll work with models, pipelines, APIs, and real product features from day one. You’ll be part of the team that takes the company from zero to one.

What You’ll Work On

- Model prototyping — experiment with CV, LLM, and generative models
- Model integration — plug open‑source models (CLIP, SAM, LLaMA, diffusion models) into our pipeline
- Data pipelines — build preprocessing flows for images, frames, and metadata
- Evaluation tooling — help design metrics for quality, consistency, and relevance
- Backend support — wrap models into simple APIs using FastAPI
- Cloud deployment — deploy experiments to AWS/GCP
- Documentation — keep experiments and findings clear and reproducible
- Collaboration — work closely with our Senior AI Engineer in weekly build cycles

What You Bring

- Strong Python + deep learning fundamentals
- Experience with CV or LLM projects (school, personal, or internship)
- Familiarity with PyTorch or TensorFlow
- Some exposure to cloud deployment (AWS/GCP)




- Curiosity, fast learning, and a builder mindset
- Clear communication and clean documentation habits

Nice to Have

- Experience with U‑Net, ResNet, CLIP, segmentation models
- LangChain / LangGraph / RAG pipelines
- FastAPI, WebSockets, Docker
- Interest in generative media (diffusion, image‑to‑video, etc.)
- Design intuition (Figma, Adobe Suite, Blender)

Why You’ll Love This Role

- You’ll be part of the founding AI team
- You’ll get direct mentorship from a senior engineer
- You’ll work on real product features, not side tasks
- You’ll see your work go into the prototype used for investors
- You’ll grow into a full‑time AI Engineer with ownership

Your First 4 Months Month 1 — Get Grounded

- Set up environment, repos, GPU, cloud
- Reproduce baseline models (CLIP, SAM, simple diffusion)
- Build a basic image/frame preprocessing pipeline
- Implement a simple LLM‑based storyline generator

Goal: Get comfortable with our stack and ship small pieces. Month 2 — Build Core Pieces

- Own a small module (scene detection, keyframe extraction, etc.)
- Build CLIP‑based scoring prototype
- Create a first version of prompt orchestration
- Wrap models into internal APIs

Goal: Start contributing to the core engine. Month 3 — Own a Subsystem

- Take full ownership of one subsystem
- Deploy a cloud‑based internal demo endpoint
- Build a easy evaluation dashboard

Goal: Operate more independently and contribute to the investor prototype. Month 4 — Full‑Time Readiness

- Optimize your subsystem
- Contribute to the end‑to‑end prototype
- Document everything cleanly
- Propose next‑phase improvements

Goal: Demonstrate readiness for full‑time ownership. Skills:- Python, Deep Learning, Computer Vision, PyTorch, Yolo and VIT

📌 AI Engineer (India)
🏢 Cutshort
📍 India