Forward Deployed Engineer - GenAI

We are seeking a skilled software engineer to develop and deploy generative AI applications, including LLM-powered features, retrieval-augmented generation (RAG) pipelines, and enterprise search solutions that move beyond prototypes to production-ready systems. This role involves close collaboration with architects, data scientists, and client engineers to translate business needs into effective AI-driven features, ensuring scalability, maintainability, and cost-efficiency.

Key Responsibilities

- Develop generative AI applications such as LLM-powered features, copilot/chat experiences, and enterprise search solutions.
- Design and implement RAG pipelines, including chunking strategies, embedding selection, hybrid retrieval, re-ranking, and GraphRAG for structured retrieval scenarios.
- Fine-tune and adapt models using techniques like LoRA and QLoRA when prompt engineering and RAG approaches are insufficient.
- Engineer and version production prompts, integrating prompt and context management within the application layer.
- Integrate LLM APIs from providers such as OpenAI, Anthropic, and Azure OpenAI, as well as open-source model endpoints, incorporating authentication, rate limiting, and cost controls.
- Instrument applications to support evaluation through output logging, quality scoring, and human-feedback loops.
- Optimize latency and token costs by implementing caching, batching, and model routing strategies.
- Translate client business requirements into detailed generative AI feature specifications.
- Communicate technical trade-offs involving cost, latency, and accuracy to non-technical product stakeholders.
- Collaborate closely with the Agentic AI Architect and Data Scientists to develop shared components.
- Document architectural decisions and prompt design for maintainability and smooth handoff.

Required Qualifications

- 4 to 8 years of software engineering experience, with 1 to 3 years focused on hands-on generative AI and LLM application development.
- Strong proficiency in Python and experience with orchestration frameworks such as LangChain or LlamaIndex.
- Familiarity with vector databases and embedding strategies, including Pinecone, Weaviate, and pgvector, as well as knowledge-graph or graph-database tools like Neo4j where applicable.
- Deep understanding of LLM failure modes such as hallucination, context window limitations, and cost blowups, along with strategies to mitigate these issues.
- Experience with model fine-tuning techniques like LoRA and QLoRA, including the use of evaluation harnesses.
- Hands-on experience with enterprise generative AI and agentic platforms such as Microsoft Azure AI Foundry, AWS Bedrock (including Strands Agents SDK), and Google Vertex AI. Familiarity with open-source frameworks like LangChain and LlamaIndex is a plus when no specific platform is mandated.
- Proven skills in API design and integration, including authentication, rate limiting, and handling streaming responses.
- Knowledge of prompt versioning and LLMOps tooling such as LangSmith or Weights & Biases.
- Strong technical writing skills, capable of documenting complex RAG architectures for non-technical stakeholders.
- Comfortable working directly with client engineers in embedded delivery settings.
- Collaborative mindset, able to work effectively with architects, data scientists, and QA teams without requiring fully specified instructions.

Preferred Qualifications and Benefits

While not explicitly listed, candidates with experience in open-source AI frameworks and those who demonstrate strong communication skills and adaptability in client-facing roles will be highly valued. The role offers the opportunity to work on cutting-edge AI technologies in a collaborative environment focused on delivering impactful, production-grade solutions.