Software Engineer - Agentic AI & Data
Posted Updated 6
views
Description
Overview
BMC is hiring a hands-on Software Engineer - Agentic AI & Data to build its next-generation agentic AI platform.
The role suits an experienced engineer with experience in shipping AI products into enterprise environments. Success is judged by reliability, controllability, cost-efficiency and customer impact.
Key Responsibilities
- Design, build and evolve agentic systems that reason, plan, execute and adapt in production.
- Take AI features from concept to production in a 0-1 product setting.
- Write and review production code (Python-first) across AI pipelines, inference services and orchestration layers.
- Treat prompt engineering, tool use, memory, evaluation and guardrails as core engineering concerns.
- Design agent frameworks that balance autonomy with determinism, observability and safety.
- Make pragmatic trade-offs across latency, cost, accuracy, scalability and maintainability.
- Integrate and operate commercial and/or open-source LLMs, including model selection, embeddings, RAG and inference optimization.
- Tackle real-world issues such as hallucinations, drift, prompt regressions and failure modes.
- Deploy and operate AI services on AWS, Azure and GCP, including secure and customer-specific enterprise deployments.
- Build scalable inference and orchestration architectures with containers, APIs and distribute systems.
- Keep the platform shippable, debuggable and supportable, and act with founder-level ownership.
Required Skills & Experience
- 4-7 years of professional software development, with significant time shipping B2B product to external customers.
- Experience in Python or Java, including designing production systems.
- Experience building, deploying and operating AI-powered products in production.
- Background in building reusable components, SDKs and APIs for AI integration
- Hands-on LLM/GenAI experience in real applications (agents, copilots, automation, decision systems).
- Deep knowledge of several of: agent frameworks and orchestration; agent harness; prompt engineering and tool-use patterns; RAG and vector search; model evaluation, feedback loops and monitoring; safety, guardrails and enterprise controls.
- Hands-on use of several of: LangGraph/LangChain, LlamaIndex, Google ADK, vector databases (Pinecone, Weaviate, FAISS, Milvus), and prompts managed as versioned, testable artifacts.
- Experience running LLMs via managed platforms (AWS SageMaker/Bedrock, Vertex AI or similar) and direct APIs (OpenAI, Anthropic).
- Experience with multi-agent systems or complex agent workflows.
- Experience commercializing AI features under enterprise constraints (security, compliance, uptime).Experience deploying LLMs at scale (Kubernetes, model serving, GPU optimization).
- Comfort with ambiguity and decision-making under incomplete information.
- Ability to manager stakeholders independently and work across domains and global teams