Position: AI Developer

Location : Argentina (Remote)

Argentina Visa required or citizen

- Design and build the AI orchestration layer that coordinates multi-step, multi-agent workflows across the SDLC (requirements - design - development - testing - deployment)
- Develop and maintain integrations with LiteLLM (or equivalent LLM gateways) for model routing, rate limiting, cost tracking, and provider abstraction
- Build and extend MCP (Model Context Protocol) servers/clients to connect agents to internal tools (Jira, GitHub, Jenkins, Confluence, AWS) and external LLM CLIs (e.g., Claude Code CLI)
- Design agent architectures with well-defined roles, tool access, guardrails, and human-in-the-loop approval gates ("Human Gate")
- Implement policy layers (e.g., Project Profiles) that let humans configure and constrain autonomous agent behavior per project
- Build backend services in .NET (C#, EF Core, REST/gRPC APIs) that power the orchestration engine, and supporting tooling/automation in Python
- Contribute to the frontend (React) for the visual, drag-and-drop pipeline builder and monitoring dashboards
- Evaluate, benchmark, and iterate on prompting strategies, agent tool-use patterns, and model selection across use cases
- Partner with product/BA stakeholders to translate SDLC pain points into agent workflows and pipeline stages
- Own reliability, observability, and security for AI-invoked actions that touch production systems (code, tickets, deployments)
- Mentor other engineers on agentic system design, LLM integration patterns, and best practices





Required Qualifications

- 8+ years of professional software development experience (Staff) or 10+ years (Senior Staff), with a track record of owning complex systems end-to-end
- Strong hands-on experience with LLM-based application development: prompt engineering, tool/function calling, RAG, agent frameworks
- Practical experience with LiteLLM or comparable LLM gateway/proxy solutions (model routing, key/quota management, observability)
- Working knowledge of MCP (Model Context Protocol) — building or integrating MCP servers/clients, or equivalent tool-calling protocols
- Experience designing and building agentic systems: multi-step reasoning, orchestration, guardrails, human-in-the-loop checkpoints
- Solid backend development skills in .NET / C# (EF Core, Web APIs, async patterns)
- Proficiency in Python for scripting, automation, tooling, or ML/LLM integration work
- Frontend development experience (React or similar) sufficient to build or contribute to internal developer tools/UIs
- Familiarity with cloud infrastructure (AWS, containers/EKS) and CI/CD tooling (Jenkins, GitHub Actions)
- Solid understanding of authentication/identity patterns (OIDC/JWT, SSO — e.g., Okta)
- Excellent system design instincts: can reason about failure modes, security boundaries, and human-oversight requirements for autonomous systems

Thanks and regards,

Mohan Sundar

M2S Tech Solutions

[email protected]

Pay: From ₹10,000.00 per hour

Work Location: Remote

📌 AI Engineer (India)
🏢 M2S Tech Solutions
📍 India