Application Deadline
August 31, 2026 (Early applications will be given preference)
Job Location
Gulshan-1, Dhaka
Number of Vacancies: 2
Who are we looking for?
We are looking for a Senior Agentic AI Engineer who understands how modern AI-native software should be designed, built, and operated in production.
This role will focus on the AI-native and agentic parts of our software systems, working closely with frontend and backend engineers. The person should be hands-on and able to make practical architecture and implementation decisions.
This role will focus on the AI-native and agentic parts of our software systems, working closely with frontend and backend engineers. The person should be hands-on and able to make practical architecture and implementation decisions.
Educational Requirements
- Bachelor’s degree in engineering is strongly preferred
- Graduates from other related fields may apply if they match the requirement
Experience Requirements
- 5+ years of software engineering experience with strong backend and system architecture fundamentals.
Overview of the Responsibilities
- Design and build agentic workflows where agents can reason, use tools, call APIs, interact with other systems, handle retries and failures, wait for human approval, and safely continue long-running processes.
- Build business automation workflows such as: Collect data –> Analyze –> Prepare report –> Send email –> Get approval –> Continue process –> Update systems.
- Build integrations using Model Context Protocol (MCP), including MCP servers, clients, tools and enterprise APIs.
- Work with MCP Apps, Generative UI and A2UI where they can improve the user experience.
- Design context, memory and workflow state. Use RAG and enterprise knowledge retrieval where needed.
- Design AI systems for better performance and lower cost using model routing, context optimization, caching, semantic caching, batching and other suitable techniques.
- Design LLM-agnostic architectures and model-routing strategies. Select the right model based on task complexity, capability, cost, latency, context size and privacy requirements. Decide when to use smaller or local models and when to use frontier models.
- Decide where AI actually adds value and where normal deterministic software is the better solution.
- Implement AI observability and evaluation, including tracing, latency, cost monitoring, failure handling and regression testing.
- Make agentic workflows production-reliable using retries, backoff, idempotency, timeouts, circuit breakers, fallbacks and safe rollbacks for long-running processes.
- Establish AI governance, security and guardrails, including data privacy, model and tool access, human-in-the-loop controls and auditability.
- Help design Agentic SDLC workflows around requirement analysis, planning, development, testing, code review and documentation.
Key Requirements
- Hands-on experience with one or more agentic platforms or AI frameworks.
- Hands-on experience designing and building production applications using LLMs, Generative AI or Agentic AI.
- Strong understanding of LLM orchestration, model routing, agentic architecture, tool/function calling, MCP, workflow orchestration, RAG, context and memory management.
- Good understanding of AI security, governance, evaluation, observability, performance and cost optimization.
- Ability to evaluate new AI technologies and decide when and how they should be used in a real software product.
- Strong programming experience in at least one language such as Python or C#.
What's in it for you?
- Competitive salary with an excellent career opportunity in a focused & stable organization.
- Most importantly, a friendly work environment with the opportunity to learn from highly skilled mentors.
How would you apply?
If you believe you are the right candidate, then send your CV to career@relisource.com with this exact line as email's subject “AIENG_30092026” by August 31, 2026.
IMPORTANT: Resume attached without above subject line will not be processed. Your CV must include your photograph.
