AI Engineer (FDE)

Job Title: AI Engineer, Forward Deployed (FDE)
Location: Remote
Experience: 5–8 years
Employment Type: Full-time
About the Role
As a Forward Deployed AI Engineer, you'll work directly with our customers to turn their hardest business problems into production-grade AI systems. This is not a back-office engineering role. You'll sit with client teams, understand their workflows and data, prototype solutions quickly, and ship them into production. You'll be part engineer, part consultant, and part product thinker, and you'll bring what you learn from the field back to our core product and platform teams.
What You'll Do
Partner with enterprise customers to scope real problems, identify where AI can deliver measurable value, and define success metrics together.
Design, build, and deploy LLM-powered applications such as RAG systems, AI agents, document intelligence pipelines, and workflow automation, integrated into customers' existing systems.
Move quickly from proof-of-concept to production, owning reliability, latency, cost, security, and evaluation along the way.
Build robust evaluation frameworks to measure model quality, catch regressions, and reduce hallucinations.
Work with customer data in messy real-world environments, including on-prem setups, legacy databases, and strict data residency or compliance requirements.
Run technical workshops, demos, and architecture reviews with both engineering teams and business stakeholders.
Feed field insights back to product and engineering, and help build reusable components, templates, and playbooks.
Travel to customer sites as needed
What We're Looking For
5+ years of Data engineering experience, with strong Python skills and solid backend fundamentals (APIs, databases, async systems).
Hands-on experience building and shipping LLM applications using models from providers like Anthropic, OpenAI, or open-source models, including prompt engineering, RAG, tool use, and agent frameworks (e.g., LangChain, LlamaIndex, or custom orchestration).
Familiarity with vector databases (Pinecone, Weaviate, pgvector, Qdrant, or similar) and embedding strategies.
Experience deploying on at least one major cloud (AWS, Azure, or GCP), plus comfort with Docker, CI/CD, and basic Kubernetes.
Strong communication skills in English, and the ability to explain technical trade-offs clearly to non-technical stakeholders, including senior leadership.
A bias for action and comfort with ambiguity. You'd rather ship a working version this week than a perfect one next quarter.
Ownership mindset: you care about whether the customer actually got value, not just whether the ticket was closed.
Nice to Have
Prior experience in a client-facing role such as solutions engineering, consulting, or a startup where you wore many hats.
Experience with fine-tuning, model serving (vLLM, TGI), or running open-source models on-prem.
Domain exposure to Indian enterprise contexts like BFSI regulations (RBI, SEBI), DPDP Act compliance, or multilingual/Indic language use cases.
Knowledge of Hindi or other regional languages, which helps with both customer rapport and Indic NLP projects.
Open-source contributions, technical blogs, or side projects in the GenAI space.

Work arrangement
Yes