Senior AI Solutions Developer
Posted 18hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
Senior AI Solutions Developer at Eastwood, developing production-grade AI systems and enforcing OWASP compliance. Leading AI solutions development while mentoring team members.
Responsibilities:
- Lead AI Solutions pre-implementation review: model selection, cost benchmarking, hosting strategy, prototype trade-offs documented before any build begins
- Architect and implement LLM pipelines: prompt engineering, RAG, structured output, tool use, multi-agent flows
- Design and build REST APIs and data pipelines connecting AI components to Knit and client systems
- Own repo-level AI configuration, shared prompt libraries, agent configs
- Conduct code reviews with written feedback; mentor Junior developers; set and document best practices
- Review SNS/SQS message contracts and integration impact before any cross-service AI merge
- Lead OWASP LLM Top 10 (2025) red-team testing on every project before production release
- Ensure STRIDE threat model is complete for every new AI system, covering data poisoning, prompt injection, model extraction, and excessive-agency risks
- Collaborate with cross-functional teams to gather requirements and propose AI-based solutions that address business needs and drive innovation.
- Stay abreast of emerging AI technologies and industry trends to identify opportunities for enhancing the organization's AI capabilities.
- Evaluate the effectiveness of AI solutions, continuously refining and optimizing them to ensure optimal performance.
- Develop comprehensive documentation for AI solutions, including technical specifications for AI features, LLM APIs, ML Libraries, vector stores, etc.
- Serve as an AI evangelist, promoting the understanding and adoption of AI technologies across the organization through presentations, workshops, and training sessions.
- Provide technical support and troubleshooting for AI implementations, ensuring the prompt resolution of issues and minimal disruption to users.
Requirements:
- Bachelor’s degree in computer science, Engineering, or a related field.
- 4+ years professional software engineering/development, with 2+ years focused on production AI/ML or LLM integration
- Deep Python fluency, i.e. FastAPI or equivalent backend frameworks for production AI services
- Hands-on LLM API experience: Anthropic Claude, OpenAI GPT-4, or equivalent — including structured output, tool use, and agentic patterns
- Solid RAG implementation: chunking strategies, vector stores (Pinecone, Weaviate, pgvector), embedding models, retrieval validation
- Document intelligence: OCR pipelines, PDF extraction (PyMuPDF, pdfplumber, AWS Textract, Docling)
- AWS services: Lambda, S3, Bedrock, SageMaker or equivalent cloud AI platform
- OWASP LLM Top 10 (2025) compliance. Can identify, mitigate, and red-team test all 10 risks in production AI systems
- OWASP ASVS Level 2 secure coding application to API design, authentication, and data handling
- STRIDE threat modeling for AI systems covering data poisoning, prompt injection, model extraction, excessive agency
- Model/API selection for choosing the right model tier for the task (cost-performance fit, not default-to-best)
- Cost-per-request benchmarking with documented analysis extrapolated to 6–12 months at projected scale
- Hosting strategy, e.g. serverless vs self-hosted decision with infrastructure cost trade-off
- Prototype trade-off report, ex. 2–3 model options tested with documented accuracy, latency, and cost results
- Strong knowledge of AI technologies, including machine learning, natural language processing, and computer vision.
- Exceptional problem-solving and analytical skills, with a proven ability to design and implement innovative solutions.
- Excellent communication and interpersonal skills, with the ability to effectively collaborate with diverse teams and convey complex technical concepts to non-technical stakeholders.
Benefits:
- Remote work Monday to Friday | 9 AM to 6 PM



















