- Company Name
- Tekshapers
- Job Title
- Senior Developer (Gemini)
- Job Description
-
Job Title: Senior Developer (Gemini)
Role Summary: Design and deliver production‑grade GenAI services on GCP using TypeScript/Node.js, Vertex AI, Gemini models, and Gemini CLI. Build, test, and ship secure microservices, prompt pipelines, and LLM‑enabled features, ensuring performance, cost efficiency, and compliance.
Expectations: • 8+ years in backend engineering with TypeScript/Node.js
• Deep expertise in language/runtime internals, async patterns, and performance profiling
• Proven experience building services on GCP (Cloud Run, Functions, GKE, Pub/Sub, Storage, Firestore/BigQuery, Secret Manager) and Vertex AI (endpoints, model registry, evals)
• Hands‑on with Gemini models (text, multimodal, function calling) and Gemini CLI for experimentation and workflow automation
• Strong prompt engineering skills: prompt design, versioning, safety guardrails, and evaluation frameworks
• Mastery of REST/GraphQL APIs, OpenAPI/Zod schema validation, automated testing (Jest/Vitest), CI/CD (Cloud Build/GitHub Actions), IaC (Terraform), Docker
• Observability: structured logging, distributed tracing, metrics, alerts
• Security fundamentals: OAuth2/JWT, IAM, secret rotation, threat modeling
Key Responsibilities: • Design, develop, and own backend services in TypeScript/Node.js (Express/NestJS).
• Build LLM‑enabled features (retrieval, tool use, structured outputs) with Vertex AI and Gemini; automate pipelines with Gemini CLI.
• Create, version, and evaluate prompts (zero/one/few‑shot), implement safety and A/B testing.
• Develop RAG and function‑calling pipelines, integrate vector stores and scalable embeddings on GCP.
• Deploy resilient APIs on Cloud Run/Functions/GKE; handle Pub/Sub events, caching, and observability.
• Establish LLM evaluation (quality, latency, toxicity, hallucination) and telemetry loops.
• Harden services for security, privacy, and compliance (PII, data residency, secret management).
• Collaborate with product, ML, and UX teams to scope, estimate, and deliver features iteratively.
• Champion performance (p95 latency), cost efficiency (tokens/req), and reliability (SLOs, error budgets).
• Mentor junior engineers on GenAI patterns, prompt engineering, and TypeScript best practices.
Required Skills: • TypeScript/Node.js (Express/NestJS) – advanced
• GCP services: Cloud Run, Functions, GKE, Pub/Sub, Storage, Firestore/BigQuery, Secret Manager
• Vertex AI (endpoints, model registry, evals), Gemini models (text, multimodal, function calling)
• Gemini CLI for experimentation and workflow automation
• Prompt engineering (design, versioning, safety, evaluation)
• API design: REST/GraphQL, OpenAPI, Zod, automated testing (Jest/Vitest)
• CI/CD: Cloud Build, GitHub Actions, IaC (Terraform), Docker
• Observability: structured logging, tracing, metrics, alerts
• Security: OAuth2/JWT, IAM least privilege, secret rotation, threat modeling
Required Education & Certifications: • Bachelor’s degree in Computer Science, Software Engineering, or related field (or equivalent professional experience).