AI Engineer

Goodfit; Chennai District; Hybrid; INR 1,400,000 - 2,000,000; Full time

Job summary

Maarga Systems in Chennai (hybrid) is seeking an AI Engineer who can write production Python, build full-stack AI applications, and understand LLM behavior to anticipate failures. You will be embedded in client engagements from day one, owning modules, delivering features, and growing into independent contributions within your first year.

This role emphasizes hands-on delivery over research, with opportunities to design agentic pipelines, integrate enterprise systems, and shape enterprise AI

Qualifications

• 2–3 years of hands-on Python development.
• Practical experience with LLM APIs (OpenAI, Anthropic Claude, Azure OpenAI, or similar).
• Working knowledge of REST APIs and at least one frontend framework (React, Streamlit, or equivalent).
• Understanding of how LLMs work: context windows, prompting, tool calling, limitations.
• Comfortable with git, code reviews, and collaborative development.
• Can explain technical decisions clearly in writing and in conversation.
• Knowledge of Agent Loop, Human In The Loop, Checkpointing, State Management, Memory, and related concepts.

Responsibilities

• Own modules and deliver features in client-engaged AI projects.
• Build agentic pipelines and enterprise integrations.
• Grow into independent contributions within the first year.

Skills

Python

LLM APIs

REST APIs

Frontend framework

LLM understanding

Git and code reviews

Technical writing

Agentic Systems

Education

B.Tech/BE Computer Science

Tools

LangChain

LangGraph

Azure/AWS

Docker

Job description

AI Engineer — Maarga Systems

Location: Chennai (hybrid)

Experience: 2–5 years

Education: B.Tech/BE Computer Science

---

About Maarga

Maarga Systems is a boutique AI consulting firm building Agentic AI solutions for large enterprises. We work at the intersection of cutting‑edge AI research and real‑world enterprise delivery — designing multi‑agent systems, evaluation infrastructure, and AI‑native applications for clients across FMCG, retail, and global enterprise sectors.

We are a small, senior team. Everyone here builds. Everyone here learns.

---

The Role

We are hiring an AI Engineer who can write production Python, build full‑stack AI applications, and understand how LLMs work well enough to know where they fail. You will be embedded in client engagements from day one — building agentic pipelines, integrating enterprise systems, and iterating on AI solutions that real users depend on.

This is not a research role, and it is not a support role. You will own modules, deliver features, and grow into independent project contributions within your first year.

---

What You Will Build

• Agentic pipelines: multi‑agent workflows with planning, tool use, memory, and autonomous decision‑making
• RAG systems: retrieval‑augmented generation with semantic search, reranking, and context management
• Full‑stack AI interfaces: lightweight frontends connected to AI backends that business users can actually use
• Evaluation frameworks: tools to measure agent accuracy, hallucination rates, and output quality over time
• Enterprise integrations: connecting AI agents to APIs, databases, Microsoft 365, and enterprise data platforms

---

What We Need From You

Must have

• 2–3 years of hands‑on Python development — clean, structured code, not just scripts
• Practical experience with LLM APIs (OpenAI, Anthropic Claude, Azure OpenAI, or similar)
• Working knowledge of REST APIs and at least one frontend framework (React, Streamlit, or equivalent)
• Understanding of how LLMs work: context windows, prompting, tool calling, limitations
• Comfortable with git, code reviews, and collaborative development
• Can explain technical decisions clearly in writing and in conversation
• Shoiuld have built AI Agentic Systems .. Agent Loop, Human In the loop, Checkpointing, State Management, Handling Prompt Injections, AI Safety, Deterministic Harness around models, Context Engineering, Memory, Constitution.md, Skills

Good to have

• Experience with LangChain, LangGraph, or any agent orchestration framework
• Familiarity with RAG: embedding models, vector databases, chunking, reranking
• Cloud basics on Azure or AWS: deploying services, storage, logging
• Exposure to evaluation approaches for non‑deterministic AI systems
• Docker and containerisation basics

---

What We Are Looking For

• Writes code that others can read and build on
• Has built something real with an LLM API — a project, a hackathon entry, a thesis component
• Understands where AI systems fail, not just where they succeed
• Takes feedback well, iterates fast, and asks good questions when stuck
• Wants to grow quickly in a hands‑on environment

A GitHub profile, thesis, or side project involving LLMs, agents, or semantic search is a strong differentiator.

---

What You Will Learn Here

• Multi‑agent system design: orchestrating cooperating LLM agents to solve complex enterprise problems
• Context engineering: managing what information agents see, when, and how
• Production AI discipline: evaluation, failure containment, observability
• Enterprise delivery: working within real client constraints — governance, integrations, timelines
• Client communication: presenting and explaining AI systems to technical and non-technical stakeholders

---

Growth Path

| Timeline | Scope |

|----------|-------|

| 0–6 months | Feature and module ownership within a guided engagement |

| 6–12 months | Full workstream ownership on a client project |

| 12–24 months | Independent contributor; internal knowledge and tool creation |

| 24+ months | Senior AI Engineer or track toward Practice Lead |

---

Compensation & Benefits

• ₹14–20L CTC per annum, based on experience and demonstrated capability
• Performance‑linked variable component
• Learning and development support: courses, certifications, relevant conferences
• Small team, high ownership, direct access to founders

---