Job Title : Senior Gen AI / AI Engineer

Job Location : Bengaluru

Exp : 7-10 years

Notice Period : immediate - 15 days

Position Overview : We are looking for an Agentic Engineer with 7–10 years of overall experience, including a strong software engineering background and recent hands-on experience in GenAI / Agentic Engineering. The role requires someone who has owned and scaled entire product engineering systems and can design, build, and deploy enterprise grade autonomous and semiautonomous AI agents that power high impact decision workflows.

Key Responsibilities

- Design and implement agentic workflows (single agent and multiagent) for complex business decisions and task automation
- Build agents that plan, reason, act, verify, and recover using tools, memory, and context graphs
- Integrate LLMs (e.g., GPT4 / GPT4o or equivalent) with enterprise systems via APIs and event driven architectures
- Own and scale end-to-end product engineering, from architecture and backend systems to deployment, observability, and production reliability
- Develop orchestration using frameworks such as LangChain, Semantic Kernel, or custom agent runtimes
- Implement guardrails: error handling, retries, fallback logic, deterministic shells around probabilistic cores
- Instrument observability and evaluation (task success, reliability, latency, safety) for agent performance
- Collaborate with product, domain SMEs,



and data teams to translate business logic into executable agent behaviours
- Ensure security, access control, compliance, and auditability across agent actions and decisions

Requirements :

Core Software Engineering (non-negotiable)

- Data structures & algorithms, system design
- Python (strong), plus one of Go/Java/TypeScript
- REST/gRPC API design, microservices architecture
- Distributed systems, concurrency, async programming
- Git, CI/CD, Docker, Kubernetes
- Unit/integration testing, TDD
- SQL + at least one NoSQL (Postgres, MongoDB, Redis)

GenAI / Agentic Inference & Deployment

- Model serving: vLLM, TensorRT-LLM, Triton Inference Server, TGI (Text Generation Inference)
- Quantization (GPTQ, AWQ, GGUF), model compression, latency/throughput optimization
- GPU infra: CUDA basics, batching strategies, KV-cache optimization
- Multi-provider LLM routing / gateway design (OpenAI, Anthropic, Azure OpenAI, Bedrock, Vertex AI)
- Cloud platforms: Azure (AI Foundry, Azure OpenAI), AWS (Bedrock, SageMaker) — GCP as a plus
- Observability: OpenTelemetry, LLM-specific tracing (LangSmith, Arize Phoenix, Langfuse)
- Cost/latency optimization, caching strategies, autoscaling for inference workloads

P.S : People with robust software engineering + Agentic background with immediate notice period will be preferred

📌 Senior AI/Gen AI Engineer (Bengaluru)
🏢 Asper.ai
📍 Bengaluru