Engineer, Principal - Bangalore, India
Originally posted 22 September 2026 by the employer — open 0 days.
About the role
This role involves leading the design and delivery of enterprise-grade agentic AI systems that bring autonomous multi-step reasoning to complex, tool-heavy workflows.
What you'll do
- Design and implement LangGraph-based agent graphs for multi-step, tool-heavy workflows.
- Build and maintain MCP (Model Context Protocol) tool wrappers to connect agents to enterprise data sources and domain-specific tooling.
- Design ontology schemas (OWL/RDF or equivalent) to capture domain-specific facts for grounding LLM reasoning.
- Build and maintain RAG pipelines and vector stores over domain documentation, including ingest, chunking strategy, embedding model selection, and retrieval quality evaluation.
- Build evaluation frameworks for non-deterministic agent outputs, including golden dataset curation, semantic correctness scoring, and regression gates.
- Lead the discovery phase with domain experts to understand complex workflows and produce skill/task maps, EDA tool interaction catalogs, and archetype classifications.
What you'll need
- 10 years in software engineering with at least 3 years focused on AI/ML systems in production.
- Demonstrated experience building agentic or multi-step LLM systems beyond demo/prototype scale that run in production, handle failures gracefully, and are tested against real domain inputs.
- Experience working in a co-design model with non-engineering domain experts.
- Strong Python experience, including async, type annotations, and production-quality code.
- Experience with LangGraph or equivalent graph-based agent frameworks (e.g., LangChain or LlamaIndex).
- Experience with MCP (Model Context Protocol) or equivalent tool-serving patterns.
- Experience with vector databases such as Chroma, Milvus, pgvector, or similar, and embedding pipelines.
- Experience with knowledge graphs or ontologies (OWL/RDF, SPARQL, or equivalent structured knowledge systems).
- Experience with CI/CD for AI systems, including pytest, golden dataset frameworks, and regression on non-deterministic outputs.
- Bachelor's degree in Computer Science, Engineering, Information Systems, or a related field.
Nice to have
- Experience with workflow orchestration engines (Temporal, Airflow, or equivalent durable execution platforms).
- Familiarity with enterprise API integration patterns: OAuth, rate limiting, retry strategies, and auth token management.
- Experience building evaluation frameworks for LLM outputs, including domain-expert-in-the-loop curation processes.
- Fine-tuning experience on domain-specific data.
- Master's degree with ML or Systems specialization.
Skills: Agentic AI systems, LLM systems, LangGraph, RAG pipelines, Python, AI/ML systems
This role has been open 0 days — well below the 67-day median for AI/ML Hardware Engineering roles.
AI/ML Hardware Engineering · AI ML Hardware
|
Open roles in category
522
|
Median days open
67 d
|
Median salary
$235k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | Qualcomm | All employers we track in this specialty (522 roles · 76 employers) |
|---|---|---|
| Open roles in this specialty | 108 | 522 |
| Open roles in the wider Software, Firmware & Systems family | 746 | 5873 · 143 employers |
| Median days open | 97 d | 67 d (+30 d vs this employer) |
| Median salary (USD postings) | — | $235k |
Skills observed across this category: Agentic AI systems, LLM systems, LangGraph, RAG pipelines, Python, AI/ML systems
Who's hiring in this category
- Qualcomm (this employer) · 107 open roles · median 97 d
- NVIDIA · 86 open roles · median 68 d
- AMD · 46 open roles · median 60 d
- Micron Technology · 32 open roles · median 40 d
- Mobileye · 25 open roles · median 54 d
- Analog Devices · 14 open roles · median 25 d
How we counted: 522 open AI/ML Hardware Engineering (AI ML Hardware) roles from 76 employers tracked in the SemiconductorJobs index, counted 22 Sept 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.