Generative AI & LLM Engineer

Trigvent Solutions; Mohali; On-site; INR 1,200,000 - 2,200,000; Full time

Job summary

Trigvent Solutions is seeking a skilled AI Software Engineer to architect, develop, and scale production-grade Generative AI applications, multi-agent systems, and chatbots. You will build advanced RAG architectures with hybrid search, re-ranking, and dynamic chunking, and fine-tune models (Llama 3, Mistral) using LoRA/QLoRA/PEFT.

You will implement robust prompt pipelines, safety guardrails, and evaluation frameworks, while delivering high-performance streaming backends with FastAPI and Redis.

Qualifications

• 1–4 years of software development experience with focus on Generative AI and LLM workflows.
• Hands-on with LangChain, LlamaIndex, Transformers, and proprietary LLM APIs (OpenAI/Claude/Gemini).
• Knowledge of embedding models, vector databases, and semantic indexing.
• Strong Python 3.10+, async/concurrency, and REST/WebSocket APIs.

Responsibilities

• Architect and scale production-grade Generative AI apps and multi-agent systems.
• Build RAG architectures with hybrid search, re-ranking, and dynamic chunking.
• Fine-tune open-weight foundation models using LoRA/QLoRA and PEFT.
• Implement prompt engineering pipelines and LLM safety/guardrails.
• Develop high-performance streaming backends with FastAPI and Redis.
• Evaluate new AI models and libraries for adoption.

Skills

Generative AI & LLM workflows

Python async & REST APIs

Production architectures

LangChain/LlamaIndex/Transformers

Prompt engineering & guardrails

Model fine-tuning & PEFT

Problem solving & system design

GPU compute acceleration

Education

B.Tech / B.E. / BCA / MCA / M.Tech in CS/SE

Tools

LangChain

LlamaIndex

Hugging Face Transformers

OpenAI API

Claude API

Gemini API

Pinecone/Weaviate/Milvus/Chroma

FastAPI

Redis

LoRA/QLoRA/PEFT

Job description

Job Description

• Architect, develop, and scale production-grade Generative AI applications, multi-agent systems, and chatbots.
• Build advanced Retrieval-Augmented Generation (RAG) architectures with hybrid search, re-ranking, and dynamic chunking.
• Fine-tune open-weight foundation models (Llama 3, Mistral) using LoRA, QLoRA, and parameter-efficient transfer learning.
• Implement robust prompt engineering pipelines, LLM guardrails (safety & hallucination mitigation), and evaluation frameworks.
• Develop high-performance streaming backend services using FastAPI, asynchronous Python, and Redis queues.
• Stay ahead of rapid advancements in the open-source AI ecosystem and evaluate emerging AI models and libraries.

Skills & Experience

• 1 to 4 years of software development experience with proven specialization in Generative AI and LLM workflows.
• Deep hands-on expertise with LangChain, LlamaIndex, Hugging Face Transformers, and proprietary LLM APIs (OpenAI, Claude, Gemini).
• Solid understanding of embedding models, vector databases (Pinecone, Weaviate, Milvus, Chroma), and semantic indexing.
• Strong programming skills in Python 3.10+, asynchronous concurrency, and RESTful/WebSocket APIs.
• Familiarity with model fine-tuning frameworks (Unsloth, PEFT, Hugging Face TRL) and GPU compute acceleration.
• Strong problem-solving skills and experience designing production architectures for generative workflows.

Qualifications

• B.Tech / B.E. / BCA / MCA / M.Tech in Computer Science, Software Engineering, or equivalent practical industry experience.