Job Title: Senior AI Engineer

Company: YoLearn.ai

Location: Noida, India (Hybrid)

Employment Type: Full-time

Experience: 2+ years in production software engineering (1+ year shipping LLM or RAG features)

About YoLearn.aiYoLearn.ai is an AI-powered learning platform built for school students across India (Classes 1–12 and JEE, NEET, and CUET aspirants), trusted by 100,000+ learners on web, Android, and iOS. Our voice-first AI tutors, interactive digital sketchpad, syllabus-aligned notes, and AI Co-Teacher tools help students learn, practise, and prepare for exams while saving teachers valuable time.

AI is the core of our product, not an add-on—backed by Pragya, India’s first K-12 LLM built on NEP 2020 and NCF 2023 across 12+ Indian languages. We are a small, focused team working with the latest models and shipping fast. Because most of our learners are under 18, safety and privacy shape everything we build.

The RoleWe're hiring a Senior AI Engineer to build the AI that students and teachers use every day. Your primary ownership is the student memory layer: how our AI remembers each student over time and uses that context to teach better. You'll also build features from prompt to production across retrieval, agents, evaluation, and deployment.

This is a small team, so your work goes live quickly—students and teachers will be using it within weeks.

What You'll Work OnPrimary Ownership: The Student Memory Layer

- Design and own long-term memory for our AI: what it remembers about each student across sessions and features.
- Build the memory pipeline: extracting, storing, updating, and retrieving memories, including summarisation, deduplication, and letting stale facts expire.
- Make memory improve learning:



personalised explanations, recall of past mistakes and weak topics, and continuity between conversations.
- Measure memory quality: recall accuracy, relevance, latency, and cost, with evals that catch regressions.
- Keep memory safe for young users: user control, deletion, and privacy by design.

Also on Your Plate

- Design, build, and ship LLM-powered features for students and teachers.
- Build and improve retrieval-augmented generation (RAG) pipelines over educational content.
- Build agentic workflows with tool calling.
- Set up evaluation: test sets, automated scoring, and regression checks for prompts and models.
- Balance quality, latency, and cost across LLM providers.
- Ship and run production Python services, including monitoring, debugging, and reliability.
- Build safety guardrails and privacy-first data handling for young users.
- Work with product and design to turn learning problems into AI features.

What You Need
- 2+ years of professional software engineering experience, with strong Python (async, typing, testing).
- At least one LLM feature in production, not just a notebook or demo—you can explain what broke and how you fixed it.
- Experience building memory or personalisation for LLM applications: conversation summarisation, long-term memory stores, user profiles, or similar. You can explain what to remember,



what to forget, and why.
- Hands-on RAG experience: chunking, embeddings, vector databases (Qdrant, pgvector, Pinecone, or similar), retrieval tuning, and reranking.
- Experience with LLM APIs and tool/function calling from at least one major provider (OpenAI, Gemini, Claude, or open models).
- Ability to build and run REST and WebSocket APIs (FastAPI or similar) backed by PostgreSQL and Redis.
- A data-driven approach to measuring AI quality: eval sets, metrics, and comparing prompts or models on data rather than gut feel.
- Solid engineering fundamentals: Git, Docker, debugging production issues, and writing explicit, maintainable code.

Nice to Have
- Skills: Agent frameworks (such as LangGraph or LangChain), voice or multimodal AI (speech, vision, OCR), or EdTech experience.
- Startup Mindset: You’ve worked at a startup or built something from zero on your own, have a strong bias to action (ship a working version, learn from real use, iterate), and take end-to-end ownership without being asked.

What We Offer
- Ownership: Own AI features end-to-end, from architecture and design to production.
- Real Impact: Your work directly helps school students across India learn and perform better in exams.
- Hard Problems: Applied AI challenges that go well beyond a basic chatbot.
- Compensation: Competitive salary based on experience.
- Work Setup: Hybrid work from our Noida office.

How to Apply
- Email [email protected] with the subject "Senior AI Engineer". Include your CV, your GitHub or portfolio, and a few lines on an LLM or RAG system you shipped: what it did, what went wrong, and how you measured it.

📌 Senior AI Engineer (Noida)
🏢 Yolearn.ai
📍 Noida