AGENTIC AI + FORWARD DEPLOYED ENGINEERING

DETAILED

Agentic AI & Forward

Deployed Engineer (FDE)

End-to-end platform engineering, agentic AI delivery, production ownership, and stakeholder enablement

Role mission: Build, deploy, integrate, and operate enterprise-grade agentic AI solutions on AWS while working directly with business, engineering, cloud, and partner teams to convert high-value use cases into reliable production outcomes.

Role familyAI Engineering / Platform Engineering / Forward Deployed EngineeringSenioritySenior Engineer / Lead Engineer (recommended)Engagement modelHands-on, client-facing, cross-functional, production accountablePrimary platformAWS-based Agentic AI PlatformEmployment detailsLocation, reporting line, employment type, and travel to be defined by hiring team

Success in this role requires equal strength in solution engineering, AI orchestration, cloud operations, production support, and stakeholder communication.

1. Role Overview The Agentic AI & Forward Deployed Engineer is responsible for translating business and operational needs into secure, scalable, and supportable agentic AI capabilities. The resource will work across the full lifecycle: discovery, architecture, rapid prototyping, application and agent development, AWS deployment, integration, production stabilization, release management, and continuous enhancement. The role combines the product mindset of a platform owner, the technical depth of an AI and cloud engineer, and the field orientation of an FDE. The resource is expected to work close to users and stakeholder teams, resolve ambiguity through hands-on delivery, and retain accountability after launch through monitoring, incident response, documentation, and knowledge transfer.

2. Key Outcomes

- Production-ready agentic AI solutions that are measurable, secure, observable, and aligned to approved business workflows.
- Faster movement from use-case discovery to pilot and production through reusable agents, tools, APIs, memory, and knowledge components.
- Stable operations across Development, QA, UAT, and Production, supported by disciplined release, incident, and environment governance.
- Clear ownership of defects, enhancements, support tickets, documentation, and stakeholder communications.
- Improved adoption and self-sufficiency through training, shadowing, runbooks, knowledge articles, and structured handover.

1. Responsibilities 3.1 Forward Deployed Engineering & Use-Case Delivery

- Partner with business users, product owners, architects, data teams, security teams, and cloud stakeholders to understand problems, constraints, workflows, and success measures.
- Convert ambiguous requirements into executable solution designs, delivery plans, prototypes, and production increments.
- Rapidly prototype agentic workflows, validate feasibility with real enterprise data and systems, and iterate based on user feedback.
- Own integration of AI solutions into existing applications, APIs, identity patterns, data platforms, and operational processes.

- Provide hands-on support during pilots, launch, hypercare, and adoption; identify friction and convert field learning into platform improvements.
- Communicate technical trade-offs, risks, dependencies, and progress in language appropriate for technical and non-technical stakeholders.

3.2 Agentic AI Engineering
- Design and implement single-agent and multi-agent systems with transparent agent roles, tool boundaries, control flow, handoffs, and termination conditions.
- Build orchestration using AWS Bedrock Agents and/or frameworks such as LangGraph and LangChain,



selecting patterns based on reliability and maintainability.
- Implement tool calling for enterprise APIs, databases, search systems, workflow services, and approved actions, with input validation and permission controls.
- Design Retrieval-Augmented Generation pipelines, including ingestion, chunking, metadata, retrieval, grounding, citation handling, and content refresh processes.
- Implement short-term and long-term memory and context layers while managing privacy, relevance, token usage, retention, and cross-session boundaries.
- Establish evaluation strategies for task success, response quality, grounding, safety, latency, and cost; investigate failures and improve prompts, tools, routing, and data.
- Apply responsible AI and security controls, including guardrails, least privilege, auditability, human-in-the-loop approvals, fallback behavior, and safe error handling.
- Evaluate and incorporate Model Context Protocol (MCP) where appropriate for standardized tool and context integration.

3.3 Full-Stack Application Engineering
- Develop responsive user experiences using React and TypeScript for agent interactions, workflow status, feedback, and administrative functions.
- Build Python-based backend services, REST APIs, microservices, and workflow orchestration components; use FastAPI when suitable.
- Integrate frontend, backend, agent services, AWS services, and enterprise platforms with robust authentication, authorization, validation, and error handling.
- Write maintainable, testable code with clear interfaces, reusable components, code reviews, automated tests, and practical engineering standards.

3.4 AWS Platform & Infrastructure Ownership
- Engineer and operate solutions using AWS Bedrock, DynamoDB, Amazon S3, AWS Lambda, Amazon ECS/Fargate, API Gateway, IAM, CloudWatch, EventBridge, and Secrets Manager.
- Manage cloud permissions, service configuration, networking dependencies, secrets, deployments, logging, alarms, and environment-specific settings.
- Diagnose issues across agents, applications, APIs, data pipelines, IAM policies, events, containers, and dependent services.
- Contribute Infrastructure as Code using Terraform where adopted and ensure repeatable, reviewable changes.
- Optimize reliability, scalability, latency, and cloud consumption through architecture reviews and operational data.

3.5 Data & Knowledge Integration
- Integrate with Databricks, SQL data stores, and enterprise data pipelines to support retrieval, analytics, workflow context, and agent actions.
- Work with structured and unstructured data, define data contracts, and validate freshness, completeness, lineage, and access requirements.
- Support Dataiku-based integration or workflow patterns when used in the workplace.
- Own knowledge-layer lifecycle activities, including source onboarding, indexing, refresh, authorization, quality checks, and troubleshooting.

3.6 DevOps, Release & Environment Management
- Use GitHub and CI/CD practices for source control, pull requests, automated validation, artifact management, and controlled deployment.
- Coordinate promotion and validation across Development, QA, UAT, and Production environments.
- Prepare release plans, deployment guides,



rollback procedures, validation evidence, stakeholder communications, and post-release checks.
- Ensure environment readiness, resolve configuration drift, and maintain release traceability and deployment discipline.

3.7 Production Support & Operational Readiness
- Own or actively support incident management, monitoring, triage, troubleshooting, restoration, root-cause analysis, and preventive actions.
- Monitor application, infrastructure, agent, tool-calling, retrieval, latency, error, and usage signals using CloudWatch and platform telemetry.
- Maintain support processes, access-management procedures, monitoring requirements, escalation paths, and recurring platform health reviews.
- Create and maintain architecture documents, runbooks, deployment guides, knowledge-base articles, operational checklists, and support documentation.
- Own support tickets, enhancement requests, and defects from intake through prioritization, resolution, validation, and closure.

3.8 Stakeholder Management & Knowledge Transfer
- Provide concise status reporting covering impact, progress, decisions, risks, dependencies, ownership, and next actions.
- Participate in AWS shadowing and enablement activities and document acquired operational knowledge for long-term ownership.
- Deliver technical walkthroughs, support training, release briefings, and structured handover sessions.

1. Technical Skills

Area

Required skills

AWS Cloud

Bedrock, DynamoDB, S3, Lambda, ECS/Fargate, API Gateway, IAM, CloudWatch, EventBridge, Secrets Manager

Frontend

React, TypeScript, API integration

Backend

Python, REST APIs, microservices, workflow orchestration

Agentic AI

Multi-agent systems, tool calling, RAG, memory and context layers

Data Platforms

Databricks, SQL, data pipelines

DevOps

GitHub, CI/CD, release management

Operations

Monitoring, incident response, troubleshooting, runbooks

Security

IAM, secrets handling, least privilege, auditability

- Bachelors degree in Computer Science, Engineering, Information Technology, or a related discipline, or equivalent practical experience.
- Strong hands-on experience delivering cloud-native applications and services on AWS in enterprise environments.
- Demonstrated experience building LLM or agentic AI solutions involving orchestration, tool calling, RAG, and context or memory management.
- Proficiency in Python and working knowledge of React and TypeScript for full-stack delivery.
- Experience with REST APIs, microservices, event-driven patterns, SQL, data pipelines, and application integration.
- Experience supporting production systems, managing incidents and defects, and promoting releases across controlled environments.
- Ability to engage directly with users and stakeholders, structure ambiguous problems, prioritize work, and communicate clear decisions.
- Strong documentation habits and willingness to own solutions beyond initial development.

1. Preferred Qualifications

- Experience with AWS Bedrock Agents, LangGraph, LangChain, MCP, or comparable agent orchestration technologies.
- Experience integrating Databricks and/or Dataiku with enterprise applications and AI workflows.
- Working knowledge of Terraform, containerized deployment on ECS/Fargate, and mature CI/CD practices.
- Experience in regulated or highly governed enterprise environments where access, audit, validation, and release evidence are required.
- Experience in consulting, field engineering, solution architecture, customer engineering, or embedded product delivery roles.

Detailed | Confidential Hiring Document

📌 Agentic AI -Deployment (Noida)
🏢 Tecnoprism
📍 Noida