Why this role
We’re building enterprise‑grade
Agentic AI platform & applications for recruitment —from sourcing and screening to interview assistance and offer orchestration. You’ll lead a small, high‑leverage team that ships fast, measures rigorously, and scales responsibly.
What you’ll do
Own delivery end‑to‑end :
backlog, execution, quality, and timelines for Agentic AI features.
Be hands‑on (30–50% coding) :
set the technical bar in Python / TypeScript; review PRs; unblock tricky problems.
Design agentic systems :
tool‑use orchestration, planning / looping, memory, safety rails, and cost / perf optimization.
Leverage LLMs smartly :
RAG, structured output, function / tool calling, multi‑model routing; evaluate build vs. buy.
Ship production ML / LLM workflows :
data pipelines, feature stores, vector indexes, retrievers, model registries.
MLOps & Observability :
automate training / inference CI / CD; monitor quality, drift, toxicity, latency, cost, and usage.
EVALs & quality :
define task‑level metrics; set up offline / online EVALs (goldens, rubrics, human‑in‑the‑loop) and guardrails.
DevOps (T‑shaped) :
own pragmatic infra with the team—GitHub Actions, containers, IaC, basic K8s; keep prod healthy.
Security & compliance :
enforce data privacy, tenancy isolation, PII handling; partner with Security for audits.
People leadership :
recruit, coach, and grow a high‑trust team; establish rituals (standups, planning, postmortems).
Stakeholder management :
partner with Product / Design / Recruitment SMEs; translate business goals into roadmaps.
What you’ve done (must‑haves)
10+ years in software / ML; 4+ years leading engineers (TL / EM) in high‑velocity product teams.
Built and operated
LLM‑powered
or ML products at scale (user‑facing or enterprise workflows).
Strong coding in
Python, Java
and
TypeScript / Node ; solid system design and API fundamentals.
Exposure to
frontend
technologies like React, Angular, Flutter
Experience on SQL databases like Postgres, MariaDB
Practical
MLOps : experiment tracking, model registries, reproducible training, feature / vectors, A / B rollouts.
LLM tooling : orchestration (LangChain / LlamaIndex / DSPy), vector DBs (pgvector / FAISS / Pinecone / Weaviate), RAG patterns, context engineering
Observability & EVALs : ML / LLM monitoring, LLM eval frameworks (RAGAS / DeepEval / OpenAI Evals), offline+online testing and human review.
Comfortable with
DevOps : GitHub Actions, Docker, basic Kubernetes, IaC (Terraform), and one major cloud (GCP / AWS / Azure).
Familiar with
AI SDLC tools : GitHub Copilot, Cursor, Claude Code, Code Llama / Codex‑style tools; test automation.
Product mindset : measure outcomes (quality, cost, speed), not just outputs; data‑driven decisions.
Nice to have
HRTech / recruitment domain (ATS / CRM, assessments, interview orchestration).
Retrieval quality tuning, prompt‑engineering at scale, policy / guardrail systems (OpenAI / Guardrails / NeMo Guardrails).
Knowledge of
multi‑agent
frameworks, graph planners, or workflow engines (Prefect / Temporal).
Experience with
privacy‑preserving ML , tenancy isolation, regionalization.
Engineering Manager • Delhi, India