Senior AI/ML Engineer

Job description

## ## Job Description: Job Title: **Senior** **AI/ML Engineer** The Purpose of This Role As a **Lead GenAI Engineer**, you will design and build the core building blocks of this platform: a provider-agnostic LLM gateway, guardrails and governance, retrieval and agentic infrastructure, evaluation harnesses, and the observability and cost-attribution systems that make GenAI viable at enterprise scale. You will work alongside a team of platform engineers, set technical direction, and raise the engineering bar through hands-on work and mentorship. This is a **platform and infrastructure role**, not an applied / prompt-engineering role. Your users are internal engineering teams and business teams; your product is the platform they build on. The Value You Deliver - You will  collaborate with partners, iterate on requirements, follow Fidelity best practices / methodologies, have thorough understanding of the technology roadmap, advancement to design / development process. - Engineering excellence is a collective set of choices we make as technologists, whether to increase code coverage or ship a product, investing in developer productivity vs saving $$$. You will consistently make culture choices that positively impact all of engineering. - You do not compromise for the sake of social acceptance and demonstrate conviction and drive to operate in highly ambiguous situations thriving in uncertainty while keeping your eyes on the long-term vision. - You lead by example and by your code — set engineering standards, review designs and PRs, and grow the capability of the wider team. - You will drive technical discussions, arbitrate, and recommend optimal path forward in a room of highly opinionated engineers and work towards ensuring platform offerings are scalable, resilient, and secure. **The Skills that are Key to this role** **Key Responsibilities** - Own foundational platform capabilities — design, build, and operate core services such as an LLM gateway / router, a normalized multi-provider API contract,  and shared SDKs that abstract away provider differences (OpenAI, Anthropic, AWS Bedrock, Google Vertex, Azure OpenAI, and self-hosted / open models). - Build guardrails and governance — implement per-tenant policies, rate limiting, PII / data-leakage controls, content safety filters, and prompt-injection defenses as reusable platform primitives. - Design retrieval and agentic infrastructure — build shared RAG components (embedding pipelines, vector store integrations, chunking / indexing services) and tool-calling / agent-orchestration infrastructure (function calling, MCP, multi-agent frameworks). - Deliver observability and cost management — instrument the platform for latency, token usage, and quality metrics; build chargeback / showback and usage-attribution so teams see and own their consumption. - Establish evaluation frameworks — create automated, repeatable evals for quality, safety, and regression, so model and prompt changes can be shipped with confidence. - Drive architecture and technical direction — make build-vs-buy decisions, define the API contracts other teams depend on, and ensure the platform is scalable, resilient, and secure. - Lead through code and mentorship — set engineering standards, review designs and PRs, and grow the capability of the wider team. - Partner with stakeholders — translate the needs of platform consumers, security, and leadership into a clear roadmap and well-specified deliverables. - Establish best practices for token optimization, context window management, prompt compression, caching strategies, and cost-efficient AI inference. Preferred Qualifications Platform Engineering experience with a strong foundation in building and operating **production backend / distributed systems** (not just notebooks or POCs). Knowledge of AI governance, security, responsible AI, and compliance requirements.  Demonstrated ability to proliferate GenAI capabilities at scale in a safe, secure and integrable way. Must Have - Deep proficiency in **Python** (and comfort with at least one of Go / Java / TypeScript for services and tooling). - Hands-on experience building **LLM-powered systems in production** — integrating LLM APIs, RAG pipelines, prompt orchestration, and / or agentic workflows. - Solid **API and platform design** skills: designing clean, versioned, provider-agnostic interfaces that other teams consume. - Practical experience with **cloud platforms** (AWS / GCP / Azure), containerization (Docker, Kubernetes), and CI/ CD. - Working knowledge of the **GenAI ecosystem** — vector databases (pgvector, Pinecone, Milvus, Weaviate), orchestration frameworks (LangGraph, LangChain, CrewAI, LlamaIndex), and evaluation tooling. - Strong grasp of **LLMOps / MLOps** concerns: observability, cost control, versioning, rollout strategies, and reliability. - A **security- and reliability-first mindset** — you think about guardrails, data governance, failure modes, and blast radius by default. - Strong understanding of Model parameters and tuning , Prompt strategies , Cost, latency, and quality optimization , Model selection and benchmarking methodologies - Demonstrated **technical leadership**: driving architecture  and delivering complex systems end-to-end. Nice-to-Haves - Experience building **internal developer platforms**, API gateways, or multi-tenant infrastructure. - Exposure to **model serving / inference optimization** (vLLM, TGI, quantization, GPU scheduling) or selfhosting open-weight models. - Familiarity with **fine-tuning / adapters (LoRA)**, embeddings model selection, and evaluation methodology. - Experience with **cost governance, chargeback / showback**, or FinOps for AI workloads. - Contributions to **open-source** GenAI tooling or frameworks. - Domain experience in a regulated industry (fintech, healthcare) where data governance is critical. How Your Work Impacts the Organization - The Technology Foundations (TF) grou