EOS RPO
AVP/AI Engineer
Experience from 5 years to 10 Years
Assistant Vice President (AVP) – Forward Deployed AI Engineer
Location: Pune, India
Department: Center of Excellence (Data Science & AI Platform Engineering)
Work Shift: 12:30 PM to 9:00 PM IST (Flexible working hours with US overlap)
About the OrganizationWe are a premier global investment management firm dedicated to delivering high-quality research, diversified investment strategies, and innovative technology solutions to institutions and private clients worldwide. Our technology teams build enterprise-grade platforms that empower investment research, quantitative modeling, risk management, and client operations.
Role OverviewWe are seeking a Forward Deployed AI Engineer (AVP) who bridges the disciplines of an AI Engineer, Solution Architect, and Product Engineer. In this role, you will be embedded directly at the intersection of business strategy and technical execution, designing and shipping production-ready Generative AI and Agentic AI applications that solve high-value investment and enterprise operational challenges.
You will contribute to our next-generation enterprise AI and data platform, ensuring models, retrieval systems, and multi-agent workflows are secure, scalable, observable, and strictly evaluated against rigorous enterprise standards.
Key ResponsibilitiesDirect Stakeholder Engagement & Solution Delivery: Partner closely with business leaders, investment professionals, and operational teams to identify high-impact AI opportunities, rapidly build prototypes, and transition them into hardened production applications.
Agentic AI & Orchestration: Architect, build, and deploy multi-agent systems, autonomous workflows, tool-calling mechanisms, and contextual memory architectures using frameworks like LangGraph, Semantic Kernel, AutoGen, and Microsoft Agent Framework.
Enterprise RAG & Data Products: Design hybrid retrieval pipelines (vector, keyword, metadata-filtered, and graph-based retrieval) over complex enterprise datasets, creating reliable, AI-ready data products.
Model Inference & Optimization: Integrate commercial model APIs (Azure OpenAI, Anthropic Claude, Google Gemini) and self-hosted open-source models (vLLM, Hugging Face), optimizing for context windows, latency, throughput, token economics, and caching.
Evaluation, Observability & Testing: Build automated LLM evaluation harnesses, continuous regression pipelines, and observability monitoring (measuring groundedness, hallucination rates, retrieval recall, and tool compliance) before and after deployment.
Platform Engineering & Security: Ensure AI solutions integrate cleanly with enterprise APIs, event-driven architectures, CI/CD pipelines, containerized environments (Docker, Kubernetes), and strict RBAC/governance standards.
Required Experience & Core Skills
Experience Level: 5–8+ years of progressive experience in software engineering, AI engineering, data engineering, or enterprise platform engineering.
Programming Mastery: Advanced proficiency in Python (including object-oriented design and asynchronous programming) and SQL.
Generative AI & Agentic Systems: Proven track record deploying production LLM applications using RAG, embeddings, prompt/context engineering, and agent frameworks (LangGraph, Semantic Kernel, AutoGen, or equivalent).
AI Platform Ecosystems: Hands-on experience developing with major cloud AI services and model providers (Azure OpenAI, Vertex AI, Anthropic Claude, OpenAI APIs, or AWS Bedrock).
Evaluation & Observability: Production experience implementing model benchmarking, retrieval evaluation, and AI telemetry using tools like OpenTelemetry, LangSmith, Langfuse, Arize, MLflow, or custom evaluation harnesses.
Software & Platform Foundations: Solid experience with REST APIs, microservices, Docker, Kubernetes, CI/CD pipelines, and secure authentication (OAuth, managed identities, RBAC).
Preferred Qualifications
Prior experience within Investment Management, Capital Markets, FinTech, or Banking technology domains.
Experience with Model Context Protocol (MCP) servers/clients, Agent-to-Agent (A2A) orchestration, and Knowledge Graphs (Neo4j, GraphRAG).
Familiarity with distributed data processing engines such as PySpark, Databricks, or cloud data warehouses (Snowflake, BigQuery).
Experience in model fine-tuning (PEFT, LoRA) and inference optimization using vLLM, TensorRT-LLM, or Triton.
Opportunity to spearhead cutting-edge Agentic AI initiatives with direct enterprise visibility and measurable business impact.
Collaborative, intellectually stimulating engineering culture working with modern distributed platforms and cloud technologies.
Comprehensive compensation structure, competitive benefits, and flexible work arrangements.