AI Platform Engineer
Spyne. Careers | spynelabs.com
Experience: 4–10 years
Location: Bangalore, India
About Spyne Spyne builds agentic squads for enterprise data transformation. A squad is a persistent team of specialized agents that learns an enterprise's data estate, does the real
migration and modeling work, checks every result against a contract, and keeps only what it proves. Every verified delivery makes the next one faster. Patterns, connectors, and verification compound into the platform, while each customer's raw data stays with them.
The team We're a small, senior team of engineers and operators who have built enterprise software, data platforms, and AI systems. We value ownership, clear thinking, and shipping work that's been verified.
Platform sits in Engineering – Platform and builds the product every squad runs on,
along with its infrastructure, security, and operations. You'll work closely with our
Forward Deployed Engineers, whose field experience shapes what we build.
The role A squad is only as good as the context it reasons from, the harness it runs in, and what it learns from verified work. On Platform, you'll build the product that delivers all three, plus the security, compliance, and DevOps foundations enterprises need before they trust it.
This is an AI engineering role with real depth in data, security, and infrastructure. You'll own major platform features end to end, working in Python, LangGraph, and modern cloud and data infrastructure.
Responsibilities:
Build core product features end to end, including the squad runtime in LangGraph:
specialized agents, shared state, checkpointing, and review by agents and people.
– Build the Delivery Harness and verification gates: dependency-graph execution,
retries, cost controls, row, aggregate, and checksum parity, and eval harnesses in CI.
– Build the Experience Graph and data pipelines, with connectors for Oracle, SAP,
mainframe, Snowflake, and Databricks, lineage, and reuse of verified work as skills.
– Build in security and compliance: SSO and RBAC, tenant isolation, encryption, audit
trails, prompt-injection guardrails, and ISO 27001, SOC 2, and DPDP controls.
– Own DevOps across cloud, VPC, on-prem, and air-gapped installs: CI/CD,
infrastructure as code, Kubernetes, observability, and incident response.
What we look
6–10 years in software engineering, with 2+ years shipping LLM apps or agents to
production.
– Deep LangGraph (or equivalent), context engineering, and eval and tracing tools
(e.g., LangSmith, Langfuse).
– Solid SQL and data engineering with Airflow or Dagster, dbt or Spark, Kafka or
CDC, and graph, vector, and search stores.
– Production Kubernetes, Terraform, CI/CD, and observability, plus hands-on cloud
security and ISO 27001 or SOC 2 controls.
– Strong Python (Go is a plus), and a record of owning product features end to end
with clear technical writing
Nice to have– Experience with data migration tooling, schema conversion, or data diffing.
– Experience packaging software for air-gapped installs.
– Familiarity with the OWASP Top 10 for LLM applications.
– Experience with knowledge graphs, MCP servers, or agent memory frameworks.
– Open-source contributions, publications, or talks in AI, data, or infrastructure.
Pay: ₹2,000,000.00 - ₹3,000,000.00 per year
Experience:
- software engineering: 6 years (Preferred)
- shipping LLM apps: 2 years (Preferred)
Work Location: In person
📌 AI Platform Engineer (India)
🏢 Kuber
📍 India