Back to open roles

ENGINEERING

AI Engineer / Senior Engineer

Noida, India Full-time

Department: Product & Engineering

Location: Noida, India

Job scope: India

Work setting: Work from office (WFO)

Purpose of the Job

Doceree is building the first proactive intelligence layer for pharma brand teams - an agentic AI system that turns clinical intent signals, campaign data, and market context into a daily brief that drives a decision in under five minutes. We are hiring AI Engineers to build the systems behind that surface.

You will design and ship production AI systems - LLM-powered pipelines, agentic workflows, retrieval over heterogeneous healthcare data, evaluation harnesses, and the orchestration layer that sits between our signal foundation and the brand managers morning brief. Youll work closely with data scientists, product, and platform engineering to take prototypes from a notebook to something a brand team relies on everyday.

Youll thrive here if you enjoy living at the seam between LLMs, traditional ML, and well-engineered backend systems - and if you care about making AI useful, reliable, and measurably better in a regulated, real-world domain.

Key Responsibilities

- Design, build, and ship production-grade LLM and agentic AI systems powering anomaly detection, contextual synthesis, next-best-action recommendations, and one-click activation across the Daily Command product surface

- Build robust retrieval-augmented generation (RAG) pipelines over Docerees clinical intent signals, campaign data, market data, and partner data - including chunking, embedding, indexing, hybrid retrieval, reranking, and grounding strategies

- Implement agent orchestration patterns (tool use, planning, multi-step reasoning, structured outputs, guardrails) using modern frameworks (e.g., LangGraph, LlamaIndex, custom orchestration) and own the latency, cost, and reliability of those agents end-to-end

- Build evaluation and observability for AI features - offline eval sets, golden datasets, LLM-as-judge harnesses, regression tests, online metrics, prompt/version tracking, and trace inspection

- Engineer prompts, system instructions, and tool interfaces as first-class artifacts - versioned, tested,



and tied to measurable business outcomes (lift, NBRx correlation, decision time, action acceptance rate)

- Integrate and benchmark multiple model providers (OpenAI, Anthropic, Google, open-source via vLLM/TGI), and make pragmatic build-vs-buy and model-routing decisions based on quality, latency, and unit economics

- Partner with data scientists to productionize ML models (classification, ranking, recommendation, embedding, forecasting) - wrapping them in services, pipelines, and APIs that downstream agents and product surfaces can consume

- Own scalable, cloud-native deployment of AI services on AWS, with strong MLOps/LLMOps hygiene: CI/CD for models and prompts, feature/vector stores, monitoring, drift detection, cost guardrails, and safe rollout

- Implement safety, privacy, and compliance controls appropriate for healthcare/HCP data - PII/PHI handling, prompt injection defenses, output validation, auditability, and human-in-the-loop where it matters

- Collaborate with product and design to translate fuzzy user problems into well-scoped AI subsystems, then prototype quickly, evaluate honestly, and harden the winner

- Stay current with advances in LLMs, agents, retrieval, evaluation, and inference infrastructure, and bring back whats genuinely useful

Qualifications - Experience, Skills & Education

- B.Tech / M.Tech / Ph.D. in Computer Science, Information Technology, Statistics, or related quantitative discipline from a Tier 1/2 institution

- 2-8 years of experience building and shipping AI/ML systems in production, with at least 2+ years focused on LLM-based or agentic AI applications (RAG, tool-using agents, copilots, structured generation, evaluation)

- Robust programming fundamentals in Python, with solid software engineering practices (typing, testing, code review, modular design, performance awareness).



Comfortable in SQL and working with cloud-native data systems

- Hands-on experience with one or more LLM/agent frameworks (LangChain/LangGraph, LlamaIndex, Haystack, Semantic Kernel, or well-reasoned custom stacks) and vector databases (pgvector, Pinecone, Weaviate, OpenSearch, FAISS, etc.)

- Practical experience with prompt engineering, structured outputs (JSON schemas / function calling), tool use, and evaluation - including building your own eval harnesses rather than relying solely on vendor dashboards

- Working knowledge of classical ML/DL libraries (scikit-learn, PyTorch, TensorFlow, Hugging Face Transformers, SpaCy, NumPy, Pandas) and when not to reach for an LLM

- Experience deploying services on AWS (Lambda, ECS/EKS, SageMaker, Bedrock) and familiarity with MLOps/LLMOps tooling (MLflow, Weights & Biases, LangSmith, Langfuse, Arize, or equivalents)

- Demonstrated ability to operate on noisy, incomplete, and sensitive real-world data, and to design systems that fail safely, observably, and recoverably

- Strong product instincts - you can push back on a half-formed requirement, scope an MVP, and know when "good enough to ship and learn" beats "perfect on a benchmark"

- Excellent written and verbal communication; able to explain agent behavior, eval results, and trade-offs to non-technical stakeholders, and to write design docs the rest of the team will actually read

Preferred Qualifications

- Experience building multi-agent or workflow-based AI systems (planner/executor patterns, critic/verifier loops, long-running agents with state)

- Familiarity with inference optimization - quantization, batching, KV cache reuse, speculative decoding, serving with vLLM/TGI/Triton, or building latency/cost-aware model routers

- Solid grounding in the math behind ML and LLMs - linear algebra, probability, statistics, and a working understanding of how transformers and embeddings actually behave

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.

📌 AI Engineer / Senior Engineer (Noida)
🏢 Doceree
📍 Noida