What are the responsibilities and job description for the GCP Gemini AI Developer position at CoSourcing Partners?
Job Title: GCP Gemini AI Developer (3–5 Years Experience)
Location: Remote / Hybrid – Chicago preferred
Employment Type: Contract / Full-Time
Reports To: GCP Technical Lead / AI Program Manager
Purpose
The GCP Gemini AI Developer will design, build, and deploy intelligent applications leveraging Google Cloud’s Gemini models and Vertex AI platform. This role exists to operationalize advanced GenAI capabilities — including natural language understanding, multimodal reasoning, and generative automation — within scalable, secure, and production-ready cloud environments.
The developer will work hands-on across data engineering, AI model orchestration, and API integration to create AI-driven business solutions that reduce manual effort, enhance decision-making, and unlock measurable value from enterprise data.
Key Performance Outcomes (6–12 Months)
Core Responsibilities
Technical Environment
Core Google Cloud Services
Programming Stack
Complementary Tools
Ideal Profile
Success Metrics
Location: Remote / Hybrid – Chicago preferred
Employment Type: Contract / Full-Time
Reports To: GCP Technical Lead / AI Program Manager
Purpose
The GCP Gemini AI Developer will design, build, and deploy intelligent applications leveraging Google Cloud’s Gemini models and Vertex AI platform. This role exists to operationalize advanced GenAI capabilities — including natural language understanding, multimodal reasoning, and generative automation — within scalable, secure, and production-ready cloud environments.
The developer will work hands-on across data engineering, AI model orchestration, and API integration to create AI-driven business solutions that reduce manual effort, enhance decision-making, and unlock measurable value from enterprise data.
Key Performance Outcomes (6–12 Months)
| Outcome | What Success Looks Like | Measurement |
| 1. Gemini-Powered Solutions Deployed | Design, develop, and deploy at least two Gemini-based AI solutions (e.g., document summarization, chat agent, or data extraction automation) using Vertex AI Gemini APIs. | Delivered to production with >90% accuracy and <2s response time. |
| 2. Scalable Cloud Architecture | Build a modular AI microservices framework using Cloud Run / Cloud Functions with integrated authentication, logging, and monitoring. | Reusable components adopted in at least 3 future use cases. |
| 3. RAG / Context-Aware Workflows | Implement Retrieval-Augmented Generation (RAG) pipelines combining Gemini BigQuery or vector databases for knowledge grounding. | Demonstrated 25% reduction in hallucination or response variance. |
| 4. Cross-Team Enablement | Partner with Data, Automation, and AppDev teams to integrate Gemini AI into existing business workflows (e.g., UiPath, Power Platform, or ServiceNow). | Minimum of 2 successful integrations with documented ROI. |
| 5. Continuous Optimization | Monitor, retrain, and improve AI models via Vertex AI pipelines and Model Monitoring. | Demonstrated 15% performance gain over baseline models. |
- Design and deploy Gemini 1.5 Pro/Flash integrations via Vertex AI and Generative AI Studio.
- Build serverless APIs and backend services for AI workflows using Cloud Run, Functions, or App Engine.
- Develop data ingestion and preprocessing pipelines using BigQuery, Dataform, and Pub/Sub.
- Apply prompt engineering and parameter tuning to improve generative model accuracy.
- Implement RAG pipelines leveraging Vertex Matching Engine or Pinecone.
- Collaborate with automation and data teams to embed AI into existing business processes.
- Maintain compliance with security, privacy, and model governance standards.
Technical Environment
Core Google Cloud Services
- Vertex AI, Generative AI Studio, Gemini API
- BigQuery, BigQuery ML, Dataform
- Cloud Run, Cloud Functions, Cloud Storage
- Pub/Sub, Secret Manager, IAM, Cloud Build
Programming Stack
- Python or TypeScript (Google Cloud SDKs, google-generativeai, aiplatform)
- FastAPI / Flask / Node.js
- LangChain / LlamaIndex for orchestration
- SQL, Pandas, and Jupyter for data prep
Complementary Tools
- Terraform (IaC)
- GitHub / GitLab CI/CD
- Vertex AI Pipelines & Model Registry
- Vector DB (Vertex Matching Engine, Pinecone, or Weaviate)
Ideal Profile
- 3–5 years hands-on GCP development experience with AI/ML exposure
- Strong working knowledge of Vertex AI, Gemini models, and RAG pipeline design
- Demonstrated ability to move AI prototypes into production
- Strong communicator, able to collaborate across automation, data, and cloud teams
- Curious problem-solver passionate about applied AI innovation
Success Metrics
- Speed to Delivery: End-to-end deployment within 8–10 weeks per use case
- Model Effectiveness: >90% accuracy or relevance rating from business stakeholders
- Scalability: Framework reused for ≥3 additional AI initiatives
- Business Impact: 25% improvement in productivity or efficiency from deployed use cases