Senior Lead Agentic Developer
Description
About Integrant
For 34 years, Integrant has designed and delivered advanced, business solutions for U.S. enterprises in biotech, medical devices, manufacturing, finance, and related highly regulated industries. Our teams are experts in traditional web and mobile applications, AI/HPC infrastructure support, data engineering, predictive AI, machine learning, and Agentic solutions. We build long-term partnerships—two-thirds of our customers choosing to partner with us 5+ years—by solving mission-critical problems with rigor, empathy, and measurable outcomes.
Role overview
Title: Agentic Developer (Enterprise Agentic Systems)
Mission: Build production-grade AI agents that run real business processes inside enterprise environments — reliably, securely, observably, and with measurable ROI.
This is not a prompt-engineering or demo-building role. The person we hire owns the full lifecycle of an agent: design, implementation, evaluation, deployment, monitoring, and iteration. They treat an agent as a distributed software system with a probabilistic component, not as a chatbot with tools attached.
Stack-agnostic by design. We use Python and TypeScript, and frameworks such as LangGraph, the OpenAI Agents SDK, Bedrock Agents, and Copilot Studio appear in our work as examples, not requirements. We hire for the underlying engineering judgment: someone who can pick up any orchestration framework in a week because they understand what every framework is abstracting.
Key responsibilities:
1. Agent design & orchestration
- Design single-agent and multi-agent systems: planner/executor, supervisor/worker, hierarchical, and event-driven patterns, choosing the simplest pattern that meets the requirement.
- Model agent workflows as explicit state machines or graphs with defined states, transitions, retries, timeouts, and human-in-the-loop checkpoints.
- Define tool contracts (schemas, idempotency, error semantics) and implement tool-calling with validation and graceful degradation.
- Implement memory strategies: short-term context management, long-term memory stores, and session persistence with clear retention rules.
2. Retrieval & knowledge (RAG at scale)
- Build and tune retrieval pipelines: chunking, embedding, hybrid search, re-ranking, metadata filtering, and query rewriting.
- Handle enterprise document realities: permissions-aware retrieval, multi-tenant isolation, freshness and sync from source systems.
- Measure retrieval quality (recall, precision, groundedness) and fix it at the pipeline level, not the prompt level.
3. Evaluation, testing & quality
- Build eval suites before shipping: golden datasets, task-level success metrics, LLM-as-judge with calibration, regression tests on every change.
- Write unit and integration tests for tools, state transitions, and guardrails; test the non-LLM 80% like any other software.
- Run offline and online evals; use A/B or shadow deployments to validate changes against production traffic.
4. Observability & operations
- Instrument agents end-to-end: traces per run, token/cost accounting, latency per step, tool-call outcomes, and failure classification.
- Define SLOs for agents (success rate, p95 latency, cost per task) and build dashboards and alerts against them.
- Operate agents in production: on-call participation, incident analysis, root-cause on model and non-model failures.
5. Security, safety & governance
- Implement defenses against prompt injection, data exfiltration, and tool misuse; enforce least-privilege for every tool and integration.
- Apply input/output guardrails, PII handling, content policies, and audit logging that satisfies enterprise compliance.
- Design human-approval gates for high-impact actions and make agent decisions explainable to auditors and business owners.
6. Enterprise integration
- Integrate agents with ERP, CRM, ITSM, document management, and identity systems via APIs, event streams, and MCP-style tool servers.
- Handle authentication and authorization on behalf of users (OAuth, service principals, delegated access) correctly.
- Build agents that survive real data: inconsistent records, partial failures, rate limits, and legacy interfaces.
7. Delivery & collaboration
- Translate business processes into agent specifications with clear success criteria and ROI measures.
- Produce architecture decision records, runbooks, and handover documentation.
- Mentor engineers transitioning into agentic development; review designs and code for agent-specific pitfalls.
Requirements
Must-have
- 8+ years of professional software engineering, with at least 1–2 years shipping LLM-based systems to production users.
- Strong in Python or TypeScript, working proficiency in the other; comfortable with async programming, API design, and typed data models (e.g., Pydantic/Zod).
- Has built at least one agent that runs a real multi-step workflow with tool calls in production, and can describe its failure modes and how they were fixed.
- Hands-on with at least one orchestration framework (LangGraph, OpenAI Agents SDK, Bedrock Agents, AutoGen, CrewAI, or similar) and able to explain what it abstracts and when to drop to raw API calls.
- Practical RAG experience: vector and hybrid search, chunking strategy, re-ranking, and retrieval evaluation.
- Experience writing evals for LLM systems and treating prompt changes as code changes with regression coverage.
- Working knowledge of observability tooling (OpenTelemetry-style tracing, LLM tracing platforms such as LangSmith, Langfuse, Arize, or equivalent).
- Understanding of LLM security threats (prompt injection, data leakage, excessive agency) and standard mitigations.
- Experience integrating with enterprise systems through REST/GraphQL APIs, webhooks, or message queues, including auth flows (OAuth 2.0, service accounts).
- Experience with workflow engines (Temporal, or Durable Functions, or Step Functions) for long-running agent processes.
- Model Context Protocol (MCP) server or client implementation.
Nice-to-have
- Cloud deployment experience (Azure, AWS, or GCP) with containers and CI/CD
- Fine-tuning or distillation experience, or structured-output and function-calling optimization at scale.
- Knowledge-graph or semantic-layer experience for grounding agents.
- Exposure to regulated environments (SOC 2, HIPAA, GDPR) and audit requirements.
- Prior consulting or client-facing delivery experience.
Benefits
- Competitive compensation package
- PTO, full medical and dental coverage, etc.
- Opportunity to travel and work onsite with U.S. customers
- In-house Technical and English training programs
- Dedicated learning time (check out our 4Plus1 Program)
- Interest free loans
- Flexible work schedules
- Perks: events, sponsored lunch, game area, rooftop hangout + more!
Skills
- Agentic AI
- AI
- API
- Api Design
- Authentication
- AutoGen
- AWS
- Azure
- CI/CD
- Cloud
- CrewAI
- CRM
- Data Engineering
- ERP
- Event Driven Architecture
- Fine Tuning
- GCP
- Gdpr
- GraphQL
- Hipaa
- HPC
- ITSM
- Langfuse
- LangGraph
- LangSmith
- LLM
- Machine Learning
- MCP
- Microsoft Copilot
- OAuth
- Observability
- OpenAI
- OpenTelemetry
- Prompt Engineering
- Pydantic
- Python
- SOC 2
- TypeScript
- Webhooks
As published by workable · 35 questions · 12 written answers
Basics
First name, Last name, Email, Headline, Phone, Address, Photo, Education, Experience, Summary, Resume, Cover letter
Short answers (7)
- What is your nationality?
- What is your Military status?
- Have you ever been employed by Integrant before? If yes, when?
- Do you have any friends/relatives working at Integrant? If yes, please specify their names.
- What is your notice period?
- Are you willing to provide Clearance/Resignation letter (Form 6)
- Linkedin Profile Link:
Pick from a list (16)
- Are you comfortable working in hybrid model?
- Are you comfortable working with US timing? (12:00 PM to 8:00 PM)
- How many years of professional software engineering experience do you have?
- How many years of hands-on experience do you have developing and deploying LLM-based applications or AI agent systems used by real users in production?
- Have you personally designed, developed, and deployed an AI agent that executes multi-step business workflows using external tools or APIs in a production environment?
- Which statement best describes your hands-on programming proficiency in Python and TypeScript?
- Which AI agent orchestration frameworks have you used hands-on to develop agentic applications?
- Which Retrieval-Augmented Generation (RAG) capabilities have you personally implemented?
- Which AI/LLM evaluation and testing practices have you implemented?
- Which observability and production monitoring capabilities have you implemented for AI agents or LLM applications?
- Which security or governance controls have you implemented in AI agent systems?
- What is your experience integrating applications or AI agents with enterprise systems using APIs and authentication mechanisms?
- Which workflow orchestration engines have you used hands-on for long-running or multi-step business processes?
- What is your hands-on experience with Model Context Protocol (MCP)?
- Which cloud platforms and DevOps technologies have you used to deploy production applications?
- Have you worked directly with enterprise clients or business stakeholders to gather requirements, design solutions, or deliver software projects?
Written answers (12)
- Do you have limitations working on certain industries? If yes, please specify.
- Briefly describe one production AI agent you have worked on, including the business process it automated, your responsibilities, the tools or systems it interacted with, and one production failure or challenge you resolved.
- Think of the ONE AI system you're proudest of that real people are using today. Briefly, what does it do and who uses it?
- Is it still running today? What is your role in keeping it running?
- Great — let's stay with that system. Walk me through it simply: is it one AI doing everything, or several parts working together?
- When a user asks it something about YOUR company's data, how does it find the right information?
- Now that it's running — when you change something, how do you check it didn't get worse before users see it?
- Tell me about a time it failed or did something wrong in production. What happened and what did you change?
- Are there actions it is NOT allowed to do on its own? Does a person ever have to approve something first?
- Roughly, what does this system cost to run each month, and what have you done to keep that down?
- Was there ever a feature for this system you decided NOT to build, or had to justify to a manager? How did you explain it?
- Last one on this system — users' personal information passes through it. How do you protect that?
Sign in to read and join the discussion.