Experience: 5-8 years

Location: Pune

Role Overview:

This role puts a senior engineer inside the customer environment to get the platform installed, integrated, and adopted. You will run cluster and platform stack installations at customer sites, integrate with whatever identity, network and application estate the customer already has, build the integrations they need, and own technical success from proof of concept through production. You are the technical bridge in both directions: translating customer requirements into solutions and turning field experience into product feedback that gets acted on. This is a build role, not a coordination role. Travel and work from customer premises, within and outside India, are mandatory.

Key Responsibilities:

- Install and configure Kubernetes clusters and the AI platform stack in customer environments, including namespace setup, registry and artifact credentials, accelerator node labelling and endpoint configuration
- Configure TLS certificates and DNS for API and console endpoints, working within the customer’s certificate authority and naming conventions
- Deploy model bundles and configure deployments, replica groups and service tiers to match the customer’s workload profile
- Validate installations end to end and hand over a working inference endpoint with documented configuration
- Integrate the platform with customer identity providers over OIDC or SAML, and configure single sign-on, group mapping and access policy
- Build integrations, reference implementations and production application code in the customer environment, including retrieval pipelines, agentic workflows and API integrations into existing systems
- Work within customer network and security constraints: tunnel endpoints and client-side routing, firewall and proxy policy, restricted egress and air-gapped conditions
- Configure the gateway-layer controls the customer requires, such as key management, quotas, rate limits and usage reporting
- Gather, analyse and translate customer technical and business requirements into implementable solutions, and understand the existing ecosystem the solution has to fit
- Run proofs of concept and pilots against defined success criteria and be honest early when an approach will not meet them
- Own technical success across the engagement lifecycle: design, integration, benchmarking, migration,



go-live and post-launch optimisation
- Run performance and cost tuning against real customer workloads and set expectations against measured numbers rather than estimates
- Troubleshoot across application, gateway, model serving, cluster, network and hardware boundaries, isolate which layer owns a problem, and escalate correctly across the vendor boundary
- Coordinate with internal engineering teams to remove blockers, file actionable defects, and advocate internally for the changes that unblock real customer outcomes
- Run technical discussions and requirement workshops with customer teams, report status, risks and dependencies to both sides, and represent the organisation professionally, including during escalations

Tools and Technologies:

Category

Tools and technologies

Deployment and orchestration

Kubernetes, RKE2, kubectl, Helm, Docker, RHEL and Linux administration, systemd, Bash

Platform installation

Helm-based installation from OCI registries, values-file configuration, namespace and secret setup, registry pull secrets and service-account keys, accelerator node labelling, endpoint and certificate configuration

Inference APIs and model operations

OpenAI-compatible chat completions and embeddings APIs, streaming responses, Python and Node SDKs, cURL, Postman, API key and quota provisioning, model bundle and profile deployment, replica and service-tier configuration, batch size and sequence-length tuning

LLM application stack

Retrieval-augmented generation, vector stores such as pgvector, Qdrant or OpenSearch k-NN, LangChain or LlamaIndex, agentic and tool-calling patterns, evaluation harnesses

Identity integration

OIDC and SAML, Keycloak, Microsoft Entra ID, Okta, Ping, single sign-on configuration, group-to-role mapping

Networking and connectivity

IPSec site-to-site tunnels, routing, DNS, load balancers, ingress controllers, TLS and PKI, firewall and proxy policy, restricted-egress and air-gapped constraints

Observability and troubleshooting





Prometheus and Grafana, OpenSearch and Fluent Bit log queries, kubectl logs, describe and events, OpenTelemetry tracing, tcpdump and packet-level diagnostics

Automation and benchmarking

Python, Bash, Ansible, Git, Terraform exposure, k6 or Locust, time to first token and tokens-per-second measurement, latency percentile reporting

Collaboration and delivery

Jira, Confluence, Slack or Teams, ServiceNow or Freshservice, Salesforce for engagement tracking

AI-assisted engineering

Agentic AI assistants used for integration code, diagnostics and customer-facing documentation

Minimum Requirements:

- Strong hands-on engineering. You write and debug the code and configuration yourself rather than specifying it for someone else
- Practical Kubernetes and Helm experience sufficient to install, configure and troubleshoot a platform product in an unfamiliar environment
- Solid Linux administration on enterprise distributions
- Working experience with AI and ML systems, LLM applications and production deployment, including retrieval pipelines and evaluation
- Understanding cloud and infrastructure concepts, CI/CD and DevOps practice
- Networking and identity fundamentals are solid enough to debug connectivity and authentication problems inside a customer’s estate
- Ability to troubleshoot across multiple layers of a solution and correctly attribute a fault
- Prior client-facing technical delivery experience across multiple engagements, with excellent communication to both technical and business stakeholders
- High tolerance for ambiguity and shifting scope, and the ability to make decisions on site without waiting for consensus
- Willingness to travel and work from customer premises, within and outside India

Preferred Requirements:

- Prior forward deployed, solutions-engineering, professional-services or consulting background at a platform or infrastructure vendor
- Domain depth in a target vertical such as banking and financial services, public sector, telecommunications or healthcare
- Enterprise security and compliance literacy, including customer security reviews, SOC 2, ISO 27001 and data-residency requirements
- Experience with air-gapped, restricted or sovereign deployments
- Experience with vendor-boundary support models where responsibility is split across organisations

📌 Field Engineer – AI/ML (India)
🏢 Emergys
📍 India