4 months ago
Bengaluru, IndiaStaff+
Responsibilities
- Build and operate an LLM control plane and gateway with provider routing, quotas, failover, token and cost tracking, caching, safety validation, and observability.
- Ship unified REST/gRPC APIs and SDKs with normalized schemas, structured outputs, access controls, and production monitoring.
- Own agent runtime capabilities including tool registries, permissions, function calling, grounding, retrieval, orchestration, state management, and long-running workflows.
- Build platform components for classical and deep-learning training, scoring, experiment tracking, packaging, drift monitoring, retraining, and tuning.
- Develop domain graphs, entity resolution, reliable ingestion pipelines, real-time context services, and hybrid graph/vector/keyword retrieval.
- Establish evaluation, governance, versioning, approvals, audit trails, SLOs, autoscaling, spend controls, and safe human-in-the-loop actioning.
- Provide templates, CLIs, sandboxes, documentation, mentoring, and best practices for MLOps and AI safety.
Requirements
- 8+ years of experience building large-scale data, ML, or platform systems.
- Strong software engineering fundamentals in API design, concurrency, and distributed systems.
- Production Python experience plus one of Java, Scala, or Go, along with microservices and API design experience.
- MLOps experience with pipelines, experiment tracking and registries, model CI/CD, A/B testing, shadow and canary deployments, and online feature computation.
- Experience with AWS, Docker, Kubernetes, multi-tenant SaaS, and performance, reliability, and cost engineering.
- Practical machine learning experience including feature engineering, training, evaluation, drift detection, and deployment for user-facing workflows.
- Experience building or operating LLM gateways or control planes, agentic systems, tool use, orchestration, human-in-the-loop workflows, safety guardrails, and online evaluation.
- Experience with knowledge graphs, GraphQL, vector search, and hybrid retrieval patterns.
- Preferred mindset includes platform-as-product thinking, developer experience focus, systems-level observability and access-control thinking, cost awareness, vendor neutrality, documentation, and mentoring.
Tech Stack
Apache AirflowApache FlinkApache KafkaApache SparkAWSDockerGoGraphQLgRPCJavaKubernetesLightGBMMLflowNeo4jPythonScalaXGBoost
Categories
About Tekion
Tekion builds an AI-native, cloud platform for automotive retail that unifies dealers, OEMs, and partners. Its products include Automotive Retail Cloud (a dealership management system for retailers), Automotive Enterprise Cloud for manufacturers, and Automotive Partner Cloud for integrations, delivered as subscription software. Privately held and headquartered in Pleasanton, California, Tekion raised private equity funding in 2024.