AI Operations (AIOps) Lead

AI Operations (AIOps) Lead

Bangalore | On -Premises | Full -Time | AI / Platform Engineering

An AI system that's live and wrong is more dangerous than one that's simply down.

A partner company is building the reliability layer for their AI systems in production. They need an AIOps Lead who owns that problem end to end. This role is sourced and placed through Away Talent Pods, a Talent -as -a -Service model built by away.center x Perhire. You work full -time for one company, embedded in their workflows and culture from day one. Not project to project. Not one of ten vendors. A real job, with the structure and stability that comes with it, based out of the away.center workspace in Koramangala.

About Perhire: We're the AI -powered talent engine behind Away Talent Pods, a TaaS model built in partnership with away.center. We source, vet, and place full -time talent into dedicated pods for companies that need a real team, not a vendor relationship. Compliance, payroll, and continuity are handled end -to -end. You focus on the work.

What you'll do

• Own the correctness and reliability of AI systems in production. Not a PoC, not a research mandate. Live systems with real downstream consequences.
  
• Design source -of -truth logic across multiple iteration and feedback cycles so the AI always acts on the latest valid brief, not stale state
  
• Build LLM -based validation pipelines that flag errors based on current instructions, with precision to avoid false positives on older briefs
  
• Design idempotent systems that handle duplicate, out -of -order, and delayed events safely. No double -processing, no silent failures.
  
• Build and own retry logic, exponential backoff, dead -letter queues, and graceful degradation patterns for high -volume AI workloads
  
• Communicate architectural decisions clearly to non -technical stakeholders, including C -level
  

Who you are

• 4 -8 years across Backend Engineering, DevOps, SRE, or Platform Engineering, with at least 1.5 -2 years of hands -on LLM or AI systems experience in production
  
• Strong PostgreSQL skills: version -aware schemas, row locking, idempotency patterns, and safe state transitions at the data layer
  
• Proven experience with event -driven systems and queuing infrastructure (SQS, RabbitMQ, Kafka, or equivalent)
  
• Failure -first thinker. DLQs, retries, circuit breakers, and graceful degradation are designed in from day one, not retrofitted.
  
• Has shipped to production. Real systems, real users, real consequences. Not demos or personal projects.
  

How we hire

• Step 1: Resume and portfolio review. We're looking at what you've shipped, not just where you've worked. Production AI systems, event pipelines, and reliability work get our attention.
  
• Step 2: Technical interview. A 30 -minute system design conversation focused on the AIOps problem space. Idempotency, context drift, validation logic. We want to hear how you think, not test your recall.
  
• Step 3: Assignment. A short take -home task built around a real problem from the role. Typically 2 -3 hours. We're looking at how you structure your thinking, not just the output.
  
• Step 4: Onboarding. Once you're in, we handle the setup. Contracts, workspace access, tool onboarding. You're ready to work from day one.
  

What you get

✅ Role matched to your profile by Perhire's AI. You're entering a scoped process built around you, not submitting into a void.

✅ A professional away.center workspace in Koramangala, Bangalore. Designed for deep work, not a home office setup.

✅ Full compliance and employer infrastructure handled end -to -end: contracts, payroll, and IP protection from day one.

✅ High ownership on a correctness -first mandate. Rare in an industry that optimises for shipping velocity over accuracy.