Gen AI Engineer - Freelance

About the Role
We aren’t looking for someone to just write basic prompt wrappers. We are building the next generation of intelligent, context-aware applications.

As a Generative AI Intern, you will work closely with our engineering team to research, prototype, and build production-ready AI features. You’ll get your hands dirty with state-of-the-art Large Language Models (LLMs), vector databases, Retrieval-Augmented Generation (RAG) pipelines, and agentic workflows.

If you are obsessed with the intersection of traditional software engineering and cognitive architectures, this is your playground.

Key Responsibilities
Prototype & Build: Develop and test AI-driven features, assistants, or internal tooling using frameworks like LangChain, LangGraph, or LlamaIndex.

Optimize RAG Pipelines: Help build and refine Retrieval-Augmented Generation systems to improve context retrieval accuracy and reduce model hallucination.

API Integration: Connect LLMs and agentic workflows to our backend APIs, databases, and third-party services (primarily in Python).

Evaluation & Benchmarking: Test, evaluate, and benchmark different models (proprietary and open-source) for latency, cost, and output quality.

Stay on the Edge: Keep up with the weekly (sometimes daily) shifts in the Gen AI landscape and propose how we can leverage new research or tools.

What We Are Looking For (Requirements)
Strong Coding Fundamentals: Proficiency in Python and standard software development practices (Git, clean code, basic API design).

AI Foundations: A solid conceptual understanding of LLMs, tokenization, embeddings, vector spaces, and prompt engineering.

Builders Mentality: You have a GitHub repo, a side project, or a demo showcasing an LLM integration you built yourself.

Problem Solvers: You don't just ask the model for an answer; you know how to debug why a retrieval failed or why an agent got stuck in an infinite loop.

Education: Pursuing or recently graduated with a degree in Computer Science, Data Science, or a related technical field—or you have an equivalent portfolio of self-taught wizardry.

Bonus Points (Nice-to-Haves)
Experience with vector databases like Pinecone, Milvus, Chroma, or pgvector.

Exposure to building stateful, multi-agent systems or autonomous loops.

Familiarity with Docker, cloud infrastructure (GCP/AWS), or backend frameworks (FastAPI/Celery).

You’ve fine-tuned an open-source model or played around with local LLM deployment (Ollama, vLLM).