Back End Engineer

Role Summary:

The Backend Engineer will build the services and integrations that connect AI services to the AI platform and to government systems: AI gateway services and model-provider adapters, policy and data-protection controls, service APIs, workflow orchestration, and integrations with identity, document and payment platforms. Model training is not required; strong backend engineering and secure integration are.

Educational Qualification:

• B.Tech./B.E., M.Tech., MCA or M.Sc. in Computer Science, Information Technology or a related discipline

  (Must have)

• Professional certification in cloud platforms or Kubernetes is desirable but not mandatory; demonstrated work, as listed under Minimum Evidence at Screening, shall carry greater weight
• Candidates with strong verifiable work (production systems or significant open-source contributions) will be considered in lieu of formal qualification

Experience:

• 4+ years of professional backend development (6+ years preferred);

  minimum 3 years building production APIs and services

• Demonstrable end-to-end ownership of at least one production backend service

  , covering design, data model, deployment and monitoring
• Experience integrating with external APIs and identity providers (OAuth 2.0, OpenID Connect, SAML) at scale
• Experience integrating LLM APIs, or building API gateways, policy engines or provider adapters, is a strong plus
• Experience with durable workflow engines or event-driven systems is a plus
• Experience with Digital India platforms such as DigiLocker or API Setu is a plus

Key Responsibilities:

API & Service Development

• Design and build APIs with clear OpenAPI contracts and versioning, design data models and optimise queries, and implement caching, queues and background jobs

AI Gateway & Platform Services

• Build and extend gateway services: adapters for self-hosted, Indian, cloud and commercial model backends behind one OpenAI-compatible API, with routing, fallbacks, circuit breakers, rate limits and caching
• Implement data-classification routing, PII detection, and usage, token and cost logging per tenant, and build service-catalogue and subscription APIs

System Integration

• Integrate AI services with government identity, document, e-office and payment systems and departmental databases, with retries, idempotency and audit logging
• Build document-intake, workflow and human-review services, including durable workflows for multi-step AI processes

Quality & Operations

• Write tested code with unit, integration and contract tests in CI, instrument services with logs, metrics and traces, and take part in on-call

Security & Privacy

• Implement authentication and authorisation, secrets handling and input validation, and apply OWASP Top 10 and OWASP Top 10 for LLM Applications controls, including defences against prompt injection at service boundaries
• Handle personal data in compliance with DPDPA 2023, including data sent to external model providers

Technical Competencies:

• Languages & Frameworks:

  Python (FastAPI, Pydantic, asyncio), and Go or Java (Spring Boot)

• APIs & Gateways:

  REST with OpenAPI, gRPC, WebSockets and server-sent events; Kong or Envoy; LiteLLM; OpenAI-compatible APIs

• Data & Messaging:

  PostgreSQL (with pgvector), Redis or Valkey; Kafka, RabbitMQ or NATS; Temporal

• Identity & Security:

  OAuth 2.0, OpenID Connect, SAML, JWT; Keycloak; HashiCorp Vault or OpenBao; Microsoft Presidio; OWASP ASVS

• Testing & Operations:

  pytest, contract testing (Pact), k6; Docker, Kubernetes, Helm; GitLab CE CI; OpenTelemetry, Prometheus, Grafana

Minimum Evidence at Screening (one required):

• A production backend service or integration built by the candidate end to end, with a verifiable link, repository or reference
• An API gateway, policy engine, provider adapter or LLM integration built by the candidate, with documented usage
• A public repository or significant open-source contribution in backend, API or developer tooling, with verifiable commit history

P.S.- Shortlisted candidates shall complete a time-bound practical assessment: a small service that fronts two model providers (one self-hosted) behind a single API, with authentication, routing by data classification, PII redaction, rate limiting, usage logging and automated tests. This shall be followed by a technical discussion of a service the candidate has built.