About Synorus

Synorus is building a next-generation ecosystem of AI-first products. Our flagship legal-AI platform LexVault is redefining legal research, drafting, knowledge retrieval, and case intelligence using domain-tuned LLMs, private RAG pipelines, and secure reasoning systems.

If you are passionate about AI, legaltech, and training high-performance models — this internship will put you on the front line of innovation.

Role Overview

We are seeking passionate AI/LLM Engineering Interns who can:

- Fine-tune LLMs for legal domain use-cases
- Train and experiment with open-source foundation models
- Work with large datasets efficiently
- Build RAG pipelines and text-processing frameworks
- Run model training workflows on Google Colab / Kaggle / Cloud GPUs

This is a hands-on engineering and research internship — you will work directly with senior founders & technical leadership.

Key Responsibilities

- Fine-tune transformer-based models (Llama, Mistral, Gemma, etc.)
- Build and preprocess legal datasets at scale
- Develop efficient inference & training pipelines
- Evaluate models for accuracy, hallucinations, and trustworthiness
- Implement RAG architectures (vector DBs + embeddings)
- Work with GPU environments (Colab/Kaggle/Cloud)
- Contribute to model improvements, prompt engineering & safety tuning

Must-Have Skills





- Strong knowledge of Python & PyTorch
- Understanding of LLMs, Transformers, Tokenization
- Hands-on experience with HuggingFace Transformers
- Familiarity with LoRA/QLoRA, PEFT training
- Data wrangling: Pandas, NumPy, tokenizers
- Ability to handle multi-GB datasets efficiently

Bonus Skills

(Not mandatory — but a robust plus)

- Experience with RAG / vector DBs (Chroma, Qdrant, LanceDB)
- Familiarity with vLLM, llama.cpp, GGUF
- Worked on summarization, Q&A; or document-AI projects
- Knowledge of legal texts (Indian laws/case-law/statutes)
- Open-source contributions or research work
-

What You Will Gain

- Real-world training on LLM fine-tuning & legal AI
- Exposure to production-grade AI pipelines
- Direct mentorship from engineering leadership
- Research + industry project portfolio
- Letter of experience + potential full-time offer
-

Ideal Candidate

- You experiment with models on weekends
- You love pushing GPUs to their limits
- You prefer research + implementation over theory alone
- You want to build AI that matters — not just demos

Location - Remote

Stipend - 5K - 10K

Skills:- Google colab, Retrieval Augmented Generation (RAG), Large Language Models (LLM) tuning, Python, PyTorch, Huggingface, Qdrant and ChromaDB

📌 AI engineer Intern (India)
🏢 Synorus
📍 India