Forward Deployed AI Engineer · 30+ production agents

Dheeraj Anikar

I build and deploy human-centered AI agents that are safe, auditable, and actually used.

I work directly with product, infra, and customer teams to take LLM systems from prototype to production — orchestration, eval, compliance, rollout. Most of my work ships into regulated environments like credit unions.

38+

AI agents live

20+

Institutions deployed

~35%

Call deflection

View Projects CV / Experience

Agent Orchestration Safety & Compliance Eval / QA Systems

What I Do in the Field

Active

LLM Product Strategy

I turn fuzzy “we should have AI” ideas into production flows: use cases, eval loops, UX, KPIs, rollout plans.

Deployed

Responsible AI Systems

Guardrails, verifier agents, policy enforcement, human-in-the-loop escalation. Keeps answers compliant and on-policy, not hallucinated.

Shipping

Scalable AI Platforms

I work with infra / integrations teams on orchestration, RAG, observability, and CI/CD so we can safely ship fast.

About Me

I’m an AI Forward Deployed Engineer at Eltropy, where I help financial institutions deploy trustworthy LLM-powered agents — chat, voice, and internal assistants that deliver value without sacrificing safety or compliance.

My work sits at the intersection of conversation design, orchestration, and control surfaces blending prompt engineering, integrations, and post-deployment evaluation to make sure intelligent systems act reliably under real-world pressure.

I’ve deployed 38+ AI agents across 20+ credit unions, building frameworks for agentic QA, retrieval-grounded reasoning, and verifier loops that keep models aligned and auditable in production.

Mission: Ship useful AI that respects constraints.