AARC Group is focused on building multi-agent, agentic AI systems for real, business-critical work. We design AI agents that plan, reason, and act, orchestrating multi-step workflows end to end, calling tools and APIs, retrieving and reasoning over documents, and returning trustworthy, structured outputs. Combining multi-agent orchestration, retrieval-augmented generation, document intelligence, and computer vision on a modern cloud, data, and AI stack (Microsoft Azure and Fabric, with a growing multi-cloud, multi-model footprint), our systems automate complex processes across AARC Group's businesses: environmental and regulatory compliance, engineering, tax, and financial advisory.
This role is based in India and can be performed from either our Noida or Hyderabad office.
Job Overview
As part of AARC's AI & Analytics team, you will lead the design, build, and operation of the secure, scalable AI systems that power automation, intelligence, and client-facing experiences across AARC. As AI Engineering Manager you will combine deep, hands-on AI engineering with people leadership, owning the technical architecture and engineering delivery, and leading and mentoring the engineers who build it.
You will architect RAG systems and production, multi-agent AI systems that plan tasks, call tools and APIs, and orchestrate real workflows. You will own agent and model design, evaluation, deployment, and monitoring, translate AARC's AI direction into robust technical architecture and delivery, and partner with data engineers, the cloud and data platform team, and business experts to ship reliable, trustworthy AI.
What You Will Do
Technical Architecture & Engineering Strategy
- Contribute to the technical and solution architecture for AARC's AI products: agentic systems, RAG, document intelligence, computer vision, and intelligent automation.
- Contribute to AARC's AI strategy and roadmap, and translate AI direction into clear technical plans, milestones, and standards.
- Make the key architecture decisions: multi-agent patterns, retrieval design, model selection, fine-tuning vs. prompting, evaluation etc.
- Set engineering standards and reusable patterns for RAG, vector databases, prompt and tool-use orchestration, guardrails, and evaluation.
Hands-On AI Engineering (senior level)
- Design autonomous and human-in-the-loop AI agents that plan, reason, and execute multi-step tasks (task decomposition, routing, reflection, retry/rollback).
- Build multi-agent patterns (planner/solver/critic, researcher/writer/reviewer) using frameworks such as LangGraph, AutoGen, CrewAI, Semantic Kernel, or custom state machines.
- Implement tool-use/function-calling so agents can query databases, retrieve documents, call APIs, trigger workflows, and return structured outputs.
- Architect end-to-end RAG pipelines: vector-database design (e.g., FAISS, pgvector, Milvus), ingestion (chunking, embeddings, hybrid search), and citations/grounding over documents such as reports, permits, and contracts.
- Lead LLM fine-tuning and optimization using parameter-efficient methods (QLoRA, PEFT, LoRA) where it adds value.
Computer Vision & Spatial AI
- Build computer vision capabilities for imagery and document use cases: object detection, segmentation, classification, and OCR over photos, drone/aerial imagery, and scanned records.
- Develop models and pipelines for LiDAR and point-cloud data, supporting site, terrain, infrastructure, and environmental use cases (e.g., feature extraction, measurement, and change detection).
- Integrate vision and spatial outputs into agentic workflows and reporting alongside language models.
Productionization, Evaluation & Observability
- Ship agents and models behind secure APIs; implement versioning, canary and A/B rollouts, and automated regression tests.
- Set up agent telemetry (traces, spans, tool-latency, token/cost), drift and outlier alerts, and safety rails.
- Define evaluation suites (task success, factuality/grounding, latency, cost, user satisfaction) and drive continuous improvement.
- Add guardrails and policies (PII masking, rate limits, cost/latency budgets, escalation) suitable for secure, compliance-aware environments.
People Management
- Manage, mentor, and grow AARC's AI engineering team; engineering standards, code reviews, and career development.
- Act as a trusted technical advisor to leadership, translating business objectives and AI direction into technical plans and solutions.
- Lead knowledge-sharing sessions and document patterns and best practices.
- Proactively raise risks, blockers, dependencies, and timelines; communicate technical tradeoffs clearly to technical and executive audiences.
Collaboration
- Partner with data engineers, the cloud and data platform team, full-stack engineers, and business and subject-matter experts to turn requirements into robust AI solutions.
- Communicate complex AI concepts clearly to non-technical stakeholders.
Required Qualifications
- Bachelor's or Master's degree in Computer Science, AI/ML, or a related field, or equivalent practical experience.
- 8+ years in AI/ML or software engineering, including substantial hands-on experience building production AI/LLM systems, and 4+ years managing or leading engineers.
- Strong programming ability (especially Python) and familiarity with ML frameworks such as PyTorch or TensorFlow and LLM orchestration frameworks such as LangChain, or comparable.
- Hands-on experience building RAG systems and/or LLM-powered agents in production.
- Experience with computer vision (e.g., object detection, segmentation, classification, or OCR) and the deep-learning frameworks behind it.
- Depth in at least one core area (advanced RAG, multi-agent orchestration, or LLM fine-tuning/adaptation such as QLoRA, PEFT, LoRA), with the ability to lead across all of them.
- Experience with vector databases (Pinecone, Azure AI or similar) and document processing (PDF parsing / OCR) at scale.
- Experience integrating AI with tools and APIs via function-calling / tool-use (databases, enterprise systems, internal services).
- Experience owning AI and technical/solution architecture for a product or platform.
- People-management experience: hiring, mentoring, setting standards, and developing engineers.
- Clear communicator able to present complex AI concepts to non-technical and executive stakeholders.
Preferred Qualifications
- Deep experience with multi-agent frameworks (LangGraph, AutoGen, CrewAI, Semantic Kernel) and/or custom agent state machines.
- Proven track record fine-tuning LLMs (QLoRA / PEFT / LoRA).
- Experience with LiDAR or point-cloud processing, photogrammetry, or geospatial / GIS data.
- Experience with LLMOps / agent observability, evaluation frameworks, and prompt management.
- Experience deploying AI workloads across cloud providers (Azure OpenAI, AWS Bedrock, GCP Vertex AI) and with secure API deployment.
- Familiarity with Microsoft Azure / Fabric, Power BI, and enterprise integrations (Monday.com, JotForm, FileMaker, SharePoint/Graph).
- Relevant certifications such as AI-102 or equivalent.
- Experience in regulated or compliance-driven industries: environmental, engineering, finance, healthcare, or legal.
Growth Opportunity
This role offers ownership of AARC's AI technical architecture and engineering delivery, and leadership of the AI engineering team. You will set the engineering standards AARC builds on and help scale production AI and automation across the Group's businesses, with a clear path to grow the function as AARC takes on AI solution work for external clients.
