Role Overview
We are hiring an entry-level AI Engineer to help build a suite of production-grade AI agents that automate real business workflows. You will work on the core AI layer — building, testing, and deploying intelligent systems that plug into existing backend APIs.
This is a hands-on engineering role focused on Large Language Models, agent orchestration, document and media understanding, and data pipelines. You will not be writing frontend or UI code — your work is the brain behind the product.
What You Will Work On
You will contribute to AI systems across a range of business problems, including:
- Conversational assistants and chatbots that answer user queries against business data.
- Document and CV understanding — extracting structure, scoring content, and ranking applicants.
- Meeting and transcript analysis — turning raw conversations into structured action items and summaries.
- Screen-capture and image understanding — labeling, annotating, and reasoning over screenshots.
- Voice and video pipelines — speech-to-text, text-to-speech, and automated content generation.
- Agentic workflows that combine multiple tools, APIs, and human-in-the-loop checkpoints.
- Evaluation and quality systems — building test sets, measuring model performance, and reducing cost and latency.
Responsibilities
- Write clean, well-tested Python code that runs in production.
- Design prompts, tool definitions, and agent flows using current LLM frameworks.
- Integrate AI services with existing REST APIs and data stores.
- Build retrieval pipelines: chunk data, generate embeddings, and query vector databases.
- Create evaluation datasets and measure each model or agent against them before shipping.
- Monitor token usage, latency, and failure modes; tune for cost and reliability.
- Document your work clearly so other engineers can pick it up.
- Collaborate with senior engineers, ask good questions, and learn fast.
- Required Skills
- Python — strong fundamentals, clean code, comfortable with virtual environments, packaging, and async basics.
- LLM APIs — hands-on experience with at least one of OpenAI, Anthropic, Gemini, or open-source models (Llama, Mistral). You should know how to call them, stream responses, and use function/tool calling.
- Prompt engineering — practical ability to design, test, and iterate on prompts for reliable output.
- REST APIs — comfortable consuming and building HTTP APIs (FastAPI or Flask is fine).
- Data handling — JSON, CSV, basic SQL, working with structured and unstructured text.
- Git — branching, pull requests, and code review hygiene.
- Problem solving — you can break a vague problem into concrete steps and ship working code without constant hand-holding.
- Communication — clear written English; you can explain what you built and why.
- Experience with an agent- orchestration framework (LangChain, LangGraph, LlamaIndex, CrewAI, or similar).
- Vector databases — pgvector, Pinecone, Weaviate, ChromaDB, or FAISS.
- Retrieval-Augmented Generation (RAG) on a real dataset.
- Speech tooling — Whisper for speech-to-text, or any TTS service.
- Basic OCR or computer vision (Tesseract, EasyOCR, or vision-capable LLMs).
- AWS — even a small personal project counts.
- Docker basics
- A public portfolio: GitHub repos, Kaggle notebooks, blog posts, or demos.
Pay: ₹25,000.00 - ₹35,000.00 per month
Work Location: In person
