AI Engineer

About the internship: Selected intern's day-to-day responsibilities include: 1. Design and build agent orchestration using LangChain the control loop that lets an LLM plan, call tools, observe results, and decide what to do next 2. Choose the right orchestration pattern for the job single-agent loop, supervisor/subagent, or graph-based multi-step flows 3. Build and maintain RAG pipelines backed by vector databases 4. Design, version, and maintain reusable 'skills' scoped, testable capabilities an agent can invoke 5. Build evaluation and observability pipelines using Langfuse tracing runs, scoring outputs, catching regressions 6. Build agent memory systems using Mem0 and Zep what gets remembered across sessions and how it's used 7. Build and maintain APIs in Python to serve agent capabilities to the rest of the product 8. Decide when a task needs human-in-the-loop review versus an automated LLM-as-judge score, and build both 9. Put guardrails at the point where the agent actually acts authorizing tool calls and validating outcomes 10. Debug multi-step and multi-agent failures by tracing runs end-to-end 11. Manage cost/latency/capability tradeoffs across models, routing tasks to the right model for the job 12. Deploy and scale agent systems on GPU infrastructure capacity, cost, and reliability 13. Maintain a clean, versioned codebase on GitHub with clear commit history and review practices. Keep the system maintainable as models, tools, and provider APIs change Who can apply: Only those candidates can apply who: are available for the work from home job/internship can start the work from home job/internship between 7th Jul'26 and 11th Aug'26 are available for duration of 6 months have relevant skills and interests * Women wanting to start/restart their career can also apply. Stipend: INR₹ 13,000 - 17,000 /month Deadline: 2026-08-07 23:59:59 Other perks: Certificate Skills required: Python, GitHub, Natural Language Processing (NLP), Database Management System (DBMS), TypeScript, Artificial intelligence, Vector Dynamics, APIs, Generative AI Development, LangChain, Generative AI Tools, Prompt Engineering, LLMOps, LLM evaluation and Python Libraries Other Requirements: 1. Experience with multi-provider model routing for cost/latency optimization 2. Familiarity with security risks specific to LLM/agent systems 3. Experience running both offline (pre-deploy) and online (production) evals 4. Experience with alternative orchestration frameworks (CrewAI, LangGraph, or similar) 5. Experience with alternative vector databases or memory systems beyond your primary stack 6. Prior experience building this kind of system from zero at an early-stage company About Company: We are Vega Visionary is an EdTech provider building the future of adaptive learning. We leverage state-of-the-art Artificial Intelligence to construct dynamic educational pathways that evolve with the user. Our technology goes beyond the standard curriculum; it analyzes unique skill gaps ('weaknesses') and automatically generates targeted content to bridge them. We are looking for bright minds to join us in our goal to disrupt the industry and change the way the world learns. Additional information: Stipend Structure: Fixed pay: ₹ 12,000 - 15,000 /month Incentive pay: ₹ 1,000 - 2,000 /month