Staff AI Engineer

OrbitShift AI; Kolkata District; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

OrbitShift AI is seeking a Staff AI Engineer to lead AI strategy and architecture across the organization. You will define system designs, model choices, and cost-aware deployments for GenAI features and agents, collaborating with product and leadership.

You will pair deep GenAI experience with backend engineering depth to ensure production-grade reliability, observability, and scalable architectures, while guiding the team on when not to use AI.

Qualifications

• B.Tech / M.Tech in Computer Science or related field from reputed institute.
• 10+ years software engineering with staff-level impact.
• 2-3+ years building GenAI applications using LLMs.
• Strong backend and distributed-systems depth.
• Experience with eval, observability, guardrails, and cost control for LLM systems.

Responsibilities

• Set AI technical strategy and multi-quarter roadmap across the org.
• Define standards, guardrails, and how we ship AI quality.
• Translate business problems into AI system designs and choose approaches.
• Drive build/buy and vendor decisions with cost and risk considerations.
• Design GenAI stack across RAG, embeddings, vector stores, agents, and routing.
• Ensure production readiness including monitoring and reliability.

Skills

Backend engineering

Distributed systems

GenAI applications

LLMs usage

Model evaluation

Communication to non-technical

Cost & latency trade-offs

Education

B.Tech / M.Tech in Computer Science

Tools

AWS

PostgreSQL

OpenSearch

Bedrock

OpenRouter

CI/CD

Job description

About OrbitShift

At OrbitShift, we are building the world's first AI-native Sales Operating System, trusted by enterprise teams to accelerate their GTM motion. Our multi-agent AI system enables actionable account insights, RFP response generation, targeted nudges, and sales content. Backed by Peak XV (formerly Sequoia Capital), Stellaris Venture Partners, and other marquee investors, we are a fast-growing, 3-year-old startup with a team from Amazon, McKinsey, IIT, and Stanford.

We are looking for a Staff AI Engineer to be our internal authority on applied AI, the person who sets technical direction on what we build with AI, not just how. This is a senior individual-contributor role for someone with staff-level impact in applied AI, backed by solid backend engineering depth.

Role Overview

You will be our go-to person for everything AI, effectively a Tech Fellow for AI within the org. You'll own the AI technical strategy, translate business problems into AI system designs, decide which models and approaches fit each use case, and own the judgment calls that make GenAI systems reliable and cost-effective in production. Just as importantly, you'll know and argue, when not to use AI.

This is a hands-on staff-level IC role with real influence over the AI roadmap and how the whole team builds with AI. You'll pair deep GenAI application experience with solid backend engineering, so your designs hold up in production and not just in a notebook. We care more about depth in fundamentals, scope of impact, and judgment than familiarity with any specific tool.

Requirements

What You’ll Do?

Set AI technical direction and strategy

• Own the AI technical strategy and multi-quarter roadmap across the org , not just execute on individual use cases, but decide where AI creates durable leverage and sequence the bets accordingly.

• Set the standards, patterns, and guardrails other engineers build against, and establish how the company evaluates, ships, and operates AI quality.

• Be the internal authority on AI: translate ambiguous business problems into concrete AI system designs, and make the call on approach, including arguing against AI/LLMs when a simpler solution wins.

• Drive org-level build/buy and model/vendor decisions with real cost and risk stakes; act as the final technical escalation point for hard AI design calls.

• Select the right model for each use case and balance output quality against cost and latency — routing, prompting strategy, retrieval design, and where fine-tuning is or isn't worth it.

• Track the field closely, papers, model releases, tooling, and turn what's changed in the last 3–6 months into concrete decisions for us.

Build GenAI systems across the stack

• Design and build across the GenAI stack, RAG, embeddings and vector stores (e.g., OpenSearch), Agentic architectures, model orchestration and routing (e.g., Amazon Bedrock, OpenRouter), and prompting, with clear opinions on where each approach and framework breaks down.

• Reason clearly about Agentic flows and long-running, autonomous agents : planning, tool use, memory/state, error recovery, and human-in-the-loop control over extended tasks, as we build agents for our own engineering workflows (including coding agents) and for client-facing use cases.

• Own production judgment: eval pipelines, observability for LLM and agent systems, guardrails, hallucination mitigation, versioning, and cost control at scale.

Build on solid engineering foundations

• Bring strong backend and distributed-systems depth to bear, trusted in any architecture review, and able to own the design of the services and data flows the AI features and agents depend on, not just build on top of them (API design, event-driven/async patterns, and the reliability, consistency, and performance trade-offs that come with distributed systems).

• Operate on a modern cloud and data stack: AWS, containerization, relational databases (PostgreSQL), and solid observability and CI/CD for production services.

Partner across the org and level up the team

• Work closely with product, design, and the founding team to align on where AI creates real leverage and navigate technical trade-offs.

• Translate complex AI concepts for non-technical audiences, represent AI direction in reviews with senior leadership, and level up the rest of the engineering team on applied AI.

What We’re Looking For?

Education

B.Tech / M.Tech in Computer Science or a related field, preferably from a reputed institute (IIT, BITS, NIT, or equivalent).

Experience

• 10+ years of software engineering experience, with staff-level impact: technical influence and standard-setting that extends well beyond your own work. We weigh scope of impact over years alone.

• A strong backend and distributed-systems foundation: the kind of depth that earns trust in any architecture review and lets you own the design of production services, not just consume them.

• 2–3+ years hands-on building GenAI applications, having used LLMs extensively across a range of use cases (RAG, agents, structured extraction, content generation, etc.).

• Strong conceptual grasp of Agentic flows and long-running agents: how they plan, use tools, manage memory/state, recover from errors, and incorporate human oversight. Hands-on experience is a plus; deep understanding is required even if you haven't shipped one in production.

• Demonstrated judgment on model selection and quality-vs-cost-vs-latency trade-offs across RAG, prompting, agents, and fine-tuning: with a clear point of view on when each fits and when it doesn't.

• Production experience with eval, observability, guardrails, and cost control for LLM systems.

• Current, active awareness of AI research and model releases, able to separate durable shifts from hype and explain why they matter for us.

• Strong communication and stakeholder skills, able to translate complex AI concepts for non-technical audiences and influence direction across the org.

Nice to Have

• Hands-on experience building or operating autonomous / long-running agents.

• LLM fine-tuning experience (LoRA/PEFT, RLHF/DPO). A plus, not a requirement.

• Background in B2B SaaS or enterprise deployments.

• Experience as the AI point person on a team or org, setting standards and levelling others up.

Benefits

Why Join Us?

• Own the AI direction at a fast-growing startup: real, enterprise-grade GenAI systems customers run in production, not demos.

• Direct line to the founding team and a seat in the decisions that shape the product and the org.

• Work alongside a world-class team: engineers and operators from Amazon, McKinsey, IIT, and Stanford.

• A fast growth curve, strong ownership, and competitive compensation with equity.