Engineering Manager

Uplers Solutions Private Limited.; Nashik District; On-site; INR 1,800,000 - 3,600,000; Full time

Benefits offered by this job

Direct report to Operations Manager

Immediate joining

Growth opportunities

Monthly KPI bonus structure

Job summary

Pet Health and Wellness Company is seeking an Engineering Manager in Nashik, India. You will own the architecture of a multi-tenant AI-native SaaS platform and the media-engine infrastructure, while leading a small team and partnering with senior engineers.

The role emphasizes architectural judgement over stack familiarity. You will drive tenant isolation as a security boundary, establish CI/CD practices, and manage delivery cadence.

Qualifications

• 5-8 years total experience with 3+ in architecture or solutioning
• Experience shipping production-grade AI-agent systems and multi-tenant SaaS architectures
• Hands-on with Node.js/Fastify, PostgreSQL, Prisma and modern cloud infra

Responsibilities

• Set architecture for multi-tenant SaaS platform and tenant isolation
• Design AI agent layer: orchestration, tool calls, context management, evaluation, HITL
• Design media-engine infrastructure: multi-agent pipelines, versioned prompts, automation workflows
• Ship code 30-40% of the time and lead code reviews for juniors
• Hire, mentor and develop junior full-stack developers
• Set engineering bar for AI-native coding and CI/CD practices
• Ensure production reliability: deployment, monitoring and uptime

Skills

Fastify

LLM

Prisma

Java

PostgreSQL

Python

Node Js

Agentic AI

multi-tenant SaaS

AI-native coding

Job description

Experience: 5.00 + years

Salary: Confidential (based on experience)

Expected Notice Period: 15 Days

Shift: (GMT+05:30) Asia/Kolkata (IST)

Opportunity Type: Office ()

Placement Type: Full Time Permanent position (Payroll and Compliance to be managed by: Pet Health and Wellness Company)

(*Note: This is a requirement for one of Uplers' client - Pet Health and Wellness Company)

What do you need for this opportunity?

Must have skills required: Fastify, LLM, Prisma, Java, PostgreSQL, Python, Node Js, Agentic AI, multi-tenant SaaS, AI-native coding

Pet Health and Wellness Company is Looking for:

Engineering Manager

Years of Experience : 5-8 years (3+ in architecture or solutioning)

Location : Nashik, Maharashtra

Reports to: Operations Manager

Manages: A team of 3 junior full-stack developers

We’re hiring an Engineering Manager to own the technology of a multi-tenant AI-native SaaS platform + a media-engine infrastructure. The current stack runs Node.js, Fastify, PostgreSQL, and Prisma - but we are hiring your architectural judgement, not your stack expertise.

About The Role

You own the architecture of the multi-tenant SaaS platform serving multiple verticals AND the AI agent layer (orchestration, tool execution, HITL approval architecture) AND the media-engine infrastructure. You report to the Operations Manager.

You lead a team of 3 junior full-stack developers and partner with the existing senior engineering talent on the team.

What You’ll Do

• Set the architecture for the multi-tenant SaaS platform (tenant isolation as a security boundary, not a filter)
• Design the AI agent layer: orchestration, tool calling, context management, evaluation, HITL approval
• Design the media-engine infrastructure: multi-agent pipelines, prompt versioning, automation workflows
• Ship code yourself 30-40% of the time (you set the bar by shipping, not by reviewing)
• Run code review for 100% of the junior dev PRs
• Hire, mentor, and develop the junior full-stack developers
• Set the engineering bar: AI-native coding (Cursor / Claude Code / Copilot), shipping cadence, code review standards
• Hold the production reliability bar: CI/CD, deploy reliability, platform uptime

Agentic AI in Your Day-to-Day

AI-native coding is already a must-have above - Cursor / Claude Code / Copilot used daily. This section is about the layer above that: agents as managed production infrastructure you own, not autocomplete you use.

Your Roles And Responsibilities

• Boilerplate, CRUD endpoints, test scaffolding, migration scripts — Agent-generated code inside your review loop, not outside it
• CI/CD diagnosis — Agent that traces the pipeline config against the last structural repo change and returns a ranked hypothesis set
• Incident first-response — Agent that opens the incident, gathers logs/metrics/deploy history, and drafts the timeline before you wake to it
• Code review triage — Agent pre-reviews PRs for convention breaches, missing tests, and tenant-scoping smells. You review the diffs it clears and the ones it cannot
• Log triage across services — Agent correlates deploy logs against the last commit and returns likely causes
• Regression test generation — Agent generates the test you decide whether the test is real or merely present
• Junior onboarding — Agent reads the codebase and generates a scoped, task-ordered first-week plan per new developer
• Documentation drift — Agent diffs the codebase against the docs repo and opens a PR when they diverge

What Stays With You, Always

• Approval architecture. Where enforcement lives and why. A client-supplied flag must never bypass it.
• Tenant isolation as a security boundary. Not a filter. The enforcement point, and the mechanism that makes it verifiable. Your agent pre-reviewer flags tenant-scoping smells — you own the boundary itself.
• Agent reliability engineering. How agents fail differently from ordinary software — hallucination, looping, confident wrong output — and how you detect, contain, and measure that in production.
• Build-vs-buy and rewrite-vs-revise calls. Standing principle: the spec is the final target, the platform is the starting point. Audit first, then revise targeted gaps.
• The 30-40% hands-on shipping rule. Agents do not set the bar. You do, by shipping.

How You Run Agents As a System

You will treat agents as managed infrastructure, not as autocomplete.

• Agent inventory. Every agent we run has a named owner, a written purpose, a versioned prompt, and an evaluation set. If it has no eval set, it does not go into production. You set the standard and you hold it.
• Prompt versioning. Prompts are version-controlled artefacts with change history. A prompt change that alters agent behaviour goes through review like a code change.
• Human-in-the-loop tiers. Tier 1 auto-executes. Tier 2 executes with logging and a reversible path. Tier 3 blocks until an authorised human approves. You own where that line sits per tenant, and you own proving it cannot be bypassed.
• Eval harness. This agent is working correctly has to be a measurable metric. You define correctness per agent and run the harness on every prompt or model change. An eval regression blocks the release.
• Cost and latency budgets. Every agent carries a token cost and a latency target. Voice agent latency currently runs 2-3s against a ~300ms target — that gap is an architecture problem you own, not a ticket you forward.
• Context engineering. Long-running agent sessions fail on context, not model capability. You decide what goes into context, what gets summarised, what gets retrieved.

What We’re Looking For (Must-Have)

• 3+ years of architecture or solutioning experience owning the technical direction of a product, within a 5-8 year total career span
• Demonstrated depth in shipping production-grade systems
• Has shipped AI agent systems to production and knows where they fail (orchestration, tool calling, context management, evaluation)
• Multi-tenant SaaS expertise: tenant isolation as a security boundary
• AI-native coding: uses Cursor / Claude Code / Copilot daily
• Cloud infrastructure at production scale (AWS: containers, managed databases, cost control)
• Has led engineers and made them better
• Ships first, refactors on the way

What We’re Looking For (Preferred)

• Compliance-constrained environments (HIPAA, PCI) and BYOC / on-prem deployment
• LLM evaluation and observability
• Early-stage startup experience (first engineering hire or founding engineer)
• Productive in Node.js, Fastify, PostgreSQL, Prisma in week one (or transferable expertise in Python, Go, or Java)

Who Excels In This Role

• Takes ownership — treats the role as their own operation ships the work end-to-end
• Proactive — sees the next move before being asked ships the fix and reports
• Domain expert — has real, hands-on depth in the discipline this role covers
• Out-of-the-box thinker — when the obvious move stops working, finds the non-obvious one quickly
• Performs under stress — triages and ships in real-time when things break
• Calibrated about AI — knows where you overrode an agent and why, and where you would never delegate

What Success Looks Like

The team ships production code daily with high quality. The architecture scales to the next 10x without a rewrite. Junior developers ramp to owning features independently within 4-6 months. The Operations Manager trusts your architectural decisions enough to step back from day-to-day.

Hiring Process

• Intro call (30 min) — mutual fit and role expectations. We give you the compensation band in this call.
• 24-hour take-home task — a real exercise from our day-to-day work
• Role deep-dive (60-90 min) — a walkthrough of real work you have done
• Team + culture conversation (45 min) — with the Operations Manager and the team you would work with
• Offer + onboarding plan

Benefits

• Direct report to the Operations Manager
• Start: immediate joining priority
• Growth path: the engineering team scales under you
• monthly KPI bonus structure (paid monthly, on outcomes) + promotion ladder