Architecture
AI Platform Architect — LLMOps, Evaluation & Governance
Architect the operating layer for governed AI: eval harnesses, monitoring, prompt/model change control, cost, drift, and release discipline.
Location
India / Global Remote
Workplace
Remote
Type
Full-time
Team
Architecture
Experience
6+ years
Compensation
Compensation discussed during process
Governed production AI
Create the bar every AI system has to clear before launch.
For enterprise clients, model choice is only one part of the problem. The harder work is evaluation, monitoring, access control, rollback, cost management, and evidence.
You will define the platform patterns Ydhya reuses across engagements so AI systems can be operated after the demo is over.
What you will do
01
Design evaluation harnesses for retrieval, groundedness, tool use, safety, latency, and cost
02
Define prompt, model, dataset, and workflow versioning practices for client deployments
03
Set observability, alerting, drift checks, audit logging, and incident response patterns
04
Create reusable architecture guidance for LLMOps, MLOps, governance, and release control
What we are looking for
01
6+ years in platform engineering, ML engineering, MLOps, LLMOps, or AI architecture
02
Deep practical experience with production deployment, monitoring, CI/CD, and reliability
03
Strong understanding of LLM evaluation, prompt/model lifecycle, governance, and security risks
04
Able to turn ambiguous enterprise risk concerns into concrete engineering standards
Apply
Send a clear note about what you have shipped.
Include the role, your resume, and links to work that show your judgment. Short, concrete notes are better than generic cover letters.