Project Engineer – AI Agents & Agentic Engineering

Company: Belsterns Technologies (OPC) Private Limited

Location: Chennai, Tamil Nadu

Work Mode: Work From Office

Employment Type: Full-Time

Experience: 2+ Years

Compensation: ₹3.5 LPA – ₹8 LPA, based on demonstrated capability, practical experience, and overall role fit

About Belsterns Technologies - https://belsterns.com/

Belsterns Technologies is an AI-native solution engineering company focused on helping organizations implement AI across real business processes.

We work across enterprise AI ecosystems such as Claude, OpenAI/ChatGPT, Microsoft Copilot, and custom-built agentic systems.

Our approach starts with the business problem and works backward to design the right combination of humans, AI models, agents, tools, data, workflows, and software.

About the Role

We are looking for a Project Engineer – AI Agents & Agentic Engineering with strong practical experience in modern AI engineering.

This is not a conventional frontend, backend, or full-stack development role with AI added as an additional skill.

We are looking for engineers who understand how to use and configure modern AI engineering environments such as:

- Claude / Claude Code
- ChatGPT / Codex
- GitHub Copilot
- Microsoft Copilot / Copilot Studio
- MCP and similar agentic platforms

The ideal candidate should be able to move beyond simply using AI to generate code and understand how to design, configure, orchestrate, evaluate, and operate AI agents and agentic systems.

Key Responsibilities

- Design and develop AI agents and agentic workflows for real business and engineering problems.
- Configure AI engineering environments for projects and repositories.
- Design project-level and repository-level instructions, context, rules, and conventions.
- Work with Claude Code, ChatGPT, Codex, GitHub Copilot, Microsoft Copilot, and comparable platforms.
- Design and implement specialised agents, sub-agents, and multi-agent workflows where appropriate.
- Integrate agents with APIs, tools, databases, repositories, knowledge sources, and external systems.
- Work with MCP, tool/function calling, retrieval, grounding, memory, and state management.
- Design human-in-the-loop workflows and appropriate approval mechanisms.
- Evaluate AI-generated outputs and agent behaviour for accuracy, reliability, security, and completeness.
- Identify hallucinations, incorrect assumptions, context failures, tool failures, and other reliability issues.
- Design testing and evaluation approaches for agentic systems.
- Apply AI throughout the software development lifecycle.
- Translate business requirements into practical AI and software architectures.
- Determine when to use AI agents, deterministic automation, conventional software, or human intervention.
- Consider security, permissions, observability, scalability, latency, token usage, cost, and production reliability.
- Work across different AI ecosystems and quickly learn new AI platforms and technologies.





Required Skills & ExperienceAgentic Engineering
- 2+ years of software engineering experience.
- Strong practical exposure to modern AI engineering tools and platforms.
- Hands-on experience with one or more of Claude, Claude Code, ChatGPT, Codex, GitHub Copilot, Microsoft Copilot, Copilot Studio, or comparable agentic platforms.
- Understanding of AI agents and agentic workflows.
- Experience with context engineering and instruction design.
- Understanding of tool/function calling.
- Experience with AI-assisted or AI-native software development.
- Ability to evaluate and validate AI-generated work.

AI / LLM Architecture

Understanding of

- LLM fundamentals
- Prompt and instruction design
- Instruction hierarchy
- Context engineering
- Tool calling
- MCP
- RAG and retrieval
- Knowledge grounding
- Memory and state
- Agent orchestration
- Sub-agents
- Multi-agent systems
- Human-in-the-loop systems
- Guardrails
- Evaluation
- Observability
- Production reliability

Software Engineering

Strong foundation in

- Programming
- APIs
- Git
- Databases
- Application architecture
- Testing
- Debugging
- Authentication and authorization
- Integrations
- Cloud environments
- Deployment

Experience with JavaScript, TypeScript, Python, React, Node.js, SQL, PostgreSQL, NoSQL, Docker, CI/CD, AWS, Azure, or GCP is an advantage.

Important Hiring Requirement

Traditional software development experience alone is not sufficient for this role.

Candidates whose experience is primarily limited to frontend, backend, full-stack, cloud, or database development and who use AI tools only occasionally for code generation may not be suitable.

Similarly, simply integrating an LLM API into an application does not automatically qualify as agentic engineering experience.

We are looking for engineers who are actively developing their capabilities in AI-native and agentic engineering.

What We Look For

We value candidates who demonstrate:

- Strong systems thinking
- Practical AI engineering experience
- Understanding of agent architecture
- Strong problem-solving ability
- Ability to work with ambiguous requirements
- Good technical judgement
- Strong software engineering fundamentals
- Ability to evaluate and validate AI outputs
- Curiosity about emerging AI technologies
- Ability to learn new AI platforms quickly
- Clear technical communication
- Ownership and accountability
- Ability to think from the business problem backward

Most importantly, you should understand how to engineer reliable systems around increasingly capable AI models.

Education

Bachelor's or Master's degree in Computer Science, Information Technology, Engineering, or a related field is preferred.





Equivalent practical engineering experience will also be considered.

Demonstrated technical capability is more important than academic credentials alone.

What You Will Work On

Depending on the project, you may work on:

- Enterprise AI agents
- Agentic business workflows
- AI-enabled applications
- Claude-based solutions
- OpenAI / ChatGPT-based solutions
- Microsoft Copilot solutions
- Copilot Studio implementations
- Custom agentic applications
- Multi-agent systems
- MCP integrations
- Knowledge and retrieval systems
- Agent memory and context architectures
- AI-native software development environments
- Agent evaluation frameworks
- AI engineering automation
- Business-system integrations
- Internal engineering agents

Selection Process1. Agentic Engineering Screening

Technical assessment covering

- AI / LLM fundamentals
- Agentic engineering
- Context and instruction design
- Tool calling
- Agent workflows
- System design
- Evaluation
- Debugging
- Software engineering fundamentals

1. Technical Interview

Practical discussion and problem-solving around:
- Agent design
- AI engineering environments
- Context engineering
- Tools and permissions
- Agent orchestration
- Evaluation
- System architecture
- Failure handling
- Production considerations

1. Final Discussion

Evaluation of:
- Systems thinking
- Engineering judgement
- Problem solving
- Ownership
- Communication
- Learning ability
- Adaptability
- Overall role fit

Why Join Belsterns?
- Work in an AI-native engineering environment
- Work hands-on with modern AI platforms
- Build real enterprise AI agents and workflows
- Work across Claude, OpenAI, Microsoft, and custom AI ecosystems
- Gain experience with production AI systems
- Work on challenging business and engineering problems
- Develop deep AI systems and solution-engineering skills
- Experiment with emerging agentic engineering approaches
- Take ownership of real projects
- Work in a rapidly evolving AI engineering setting

How to Apply Interested candidates can share their updated resume along with relevant project details, GitHub/portfolio links, or examples of AI/agentic engineering work.

Careers: [email protected]

Direct Contact: [email protected]

Subject: Application – Project Engineer – AI Agents & Agentic Engineering

Candidates with demonstrable hands-on experience in AI agents, agentic workflows, Claude Code, Codex, Copilot, MCP, AI-native engineering, or similar technologies are especially encouraged to apply.

Build the Next Generation of Software Engineering With Us

If you have moved beyond simply using AI as a coding assistant and are interested in designing systems where humans, AI models, agents, tools, data, and software work together, we would like to hear from you.

Join Belsterns Technologies as a Project Engineer – AI Agents & Agentic Engineering.

Pay: ₹300,000.00 - ₹800,000.00 per year

Benefits

- Health insurance
- Provident Fund

Work Location: In person

📌 PROJECT ENGINEER – AI AGENTS & AGENTIC ENGINEERING (Chennai)
🏢 Belsterns Technologies
📍 Chennai