Job Description

We are looking for a hands-on, consultative AI Lead to sit at the front line of client engagements and turn AI ambition into working software. You will run discovery with clients, shape priority use cases into buildable solutions, and then personally build the prototypes, agents, and phase-0 MVPs that create client conviction and accelerate deal conversion.

This is a forward deployed role: you work shoulder-to-shoulder with clients and the account team rather than behind a delivery firewall. Success is measured by how quickly a vague business problem becomes a demonstrable AI capability with a credible value hypothesis.

It sits at the intersection of AI engineering, solution design, and business value — and is best suited to a strong builder who is comfortable in ambiguous, rapid-moving sales cycles and enjoys being the technical voice in the room.

Roles & Responsibilities

Discovery and use-case shaping

- Run early-stage discovery sessions and working groups with clients and account teams.
- Identify pain points, value levers, and high-impact AI opportunities across the client's process landscape.
- Translate vague or complex client questions into clear solution hypotheses, initial architectural logic, and a prioritised use-case backlog.
- Build AI scenarios and "before/after" narratives that illustrate meaningful business transformation, with measurable outcomes attached.

Build and prove
- Design AI solutions end to end for a given use case — LLM choice, prompt and context design, data sources, retrieval, tools, integrations, and a usable front end — working with the AI Architect on platform-level decisions.
- Build prototypes, storyboards, demos, accelerators, and reusable agentic blueprints at pursuit speed.
- Engineer and validate AI agents for reliability, reasoning quality, safety, and performance; stand up evaluation harnesses so claims are evidence-backed rather than anecdotal.
- Lead PoCs, pilots, and phase-0 MVPs across industries, owning scope, timeline, and the technical outcome.

Client-facing content and conviction
- Create client-ready material: solution and architecture decks, demo storylines, value frameworks, effort estimates, and inputs into proposals and SoWs.
- Lead technical walkthroughs, deep-dives, and competitive differentiation discussions with client engineering and business stakeholders.




- Refine solution narratives based on client feedback, objections, and market conditions.

Continuity and uplift
- Own a clean handover from pursuit to delivery — architecture intent, code, decisions, risks, and open questions documented and transferred.
- Contribute reusable assets back to the practice: accelerators, agent patterns, eval suites, reference implementations.
- Mentor 2–5 engineers on agentic engineering practice and raise the bar on prototype quality.

Must-have skills and experience

Agentic AI engineering

- Hands-on working knowledge of defining and configuring prompts, instructions, tools, reasoning strategies, guardrails, memory, and orchestration in AI agent development.
- Practical experience with multi-step and multi-agent workflows, tool/function calling, and failure handling (retries, fallbacks, human-in-the-loop).

RAG and context engineering

- Strong experience with RAG pipelines: chunking and indexing strategy, embeddings, vector databases, hybrid and re-ranked retrieval, evaluation of retrieval quality.
- Working understanding of tokenization, context-window management, and cost/latency trade-offs.

Engineering craft

- Hands-on experience creating and maintaining Python libraries; working command of LangChain (or equivalent orchestration frameworks), Hugging Face, OpenAI/Anthropic APIs, and locally hosted open models.
- Very strong experience developing RESTful APIs in Python with FastAPI (or similar) and integrating third-party services, UI components, and APIs.
- Hands-on Docker-based deployment and disciplined use of Git/GitHub for source control, branching, and review — mandatory.
- Enough front-end capability (React/Streamlit/Gradio or equivalent) to make a demo feel like a product.

Model provider and hyperscaler GenAI platform know-how

- Working familiarity across the major frontier-model providers and hyperscaler AI platforms, with genuine in-depth, hands-on depth in at least one of the following:
- Anthropic — Claude model family, Claude API, tool use,



extended thinking, Claude Agent SDK, MCP (Model Context Protocol) for tool and data connectivity, Claude on Amazon Bedrock / Vertex AI.
- OpenAI — GPT model family, Responses/Assistants APIs, Agents SDK, function calling, structured outputs, fine-tuning and batch APIs.
- AWS — Amazon Bedrock (multi-model access, Knowledge Bases, Guardrails, Agents/AgentCore), SageMaker, and the surrounding data and security services.
- Google — Gemini model family, Vertex AI (Model Garden, grounding, agent tooling), and Gemini enterprise offerings.
- Microsoft Azure — Azure OpenAI / Azure AI Foundry, AI Agent Service, AI Search for retrieval, and Azure-native integration patterns.

- Able to explain why a given provider or platform fits a client's constraints — model capability, context window, latency, cost per interaction, data residency and privacy posture, enterprise controls, existing cloud commitments — rather than defaulting to a personal favourite.
- Comfortable building the same use case on more than one stack, and keeping solutions reasonably provider-portable so a client is not architecturally locked in by a prototype.

Consultative ability

- Demonstrated ability to turn ambiguous business problems into crisp AI use cases with clear value hypotheses and measurable outcomes.
- Strong communication and articulation skills; able to hold a technical room and a business room in the same meeting.
- Hands-on prototyping mindset — able to quickly build functional demos, agent workflows, and LLM-powered interactions that accelerate client conviction.

Good to have

- Prior pre-sales, consulting, or forward-deployed engineering exposure.
- Depth in one or two industry domains (e.g. BFSI, healthcare/life sciences, retail, manufacturing, telecom).
- Certification on a hyperscaler AI/ML track (AWS, Azure or Google), or equivalent demonstrable platform depth.
- Familiarity with modern data platforms and their GenAI surfaces (Databricks, Snowflake Cortex, lakehouse patterns).
- Experience with self-hosted / open-weight model deployment (vLLM, Ollama, Bedrock or Vertex custom endpoints) for clients with strict data-boundary requirements.
- Exposure to LLM observability and evaluation tooling, and to responsible-AI and data-privacy considerations in regulated environments.

📌 Artificial Intelligence Engineer (Noida)
🏢 Impetus Technologies
📍 Noida