Job Description: ML/AI Engineer – Platform / Agentic / LLM
Location:
Bengaluru / Mumbai
Work Mode:
Onsite
Experience:
5–10 years
About the Role
We are hiring a
ML/AI Engineer – Platform / Agentic / LLM
who has
hands-on experience building ML/AI platforms from scratch
not just extending existing systems.
You will architect and develop the core Intelligence Platform: agentic workflows, LLM orchestration, retrieval pipelines, evaluation layers, and foundational services that future AI applications will be built on.
This is a role for engineers who have designed
new systems end-to-end
, made major architectural decisions, and shipped complex AI/LLM platforms in fast-paced product or startup environments.
Key Outcomes
Build the
AI/ML platform from scratch
— core engine, services, pipelines, orchestration, and infrastructure.
Create scalable
agentic workflows
and multi-step reasoning systems.
Design and ship
RAG architectures
, embeddings pipelines, and vector retrieval systems.
Build reusable platform components that power multiple internal and external applications.
Responsibilities
Architect and implement the
entire platform lifecycle
, from blank-sheet design → POC → scalable production system.
Develop
LLM orchestration layers
, task routers, executor engines, and evaluation frameworks.
Build
RAG systems
: chunking, vector schemas, scoring, retrieval optimization, and evaluation.
Own embedding pipelines and integrate vector databases such as Pinecone, Weaviate, FAISS, Chroma.
Implement
agentic pipelines
and multi-tool workflows that operate autonomously.
Make core architectural decisions around microservices, APIs, observability, scaling, and reliability.
Collaborate with product and leadership teams to evolve the platform’s roadmap.
Write clean, production-grade code with strong testing, monitoring, and documentation.
Required Skills & Experience
5–10 years overall engineering experience, with 3–5 years
in ML/AI or platform engineering.Mand
a
tory: Prov
en experience build
ing large-scale ML/AI/LLM platforms from scratch (not
only PoCs, not only model experimentation).Stron
g Python engineering skills async, distributed systems, concurrency, API design.Deep
experience with:LLMs
(OpenAI, Anthropic, Llama, etc.)RAG s
ystems & vector retrievalEmbed
dings, vector DBs (FAISS, Pinecone, Weaviate, Chroma)Orche
stration frameworks (LangChain, LlamaIndex, custom architectures)Stron
g understanding of platform principles: service design, data pipelines, load distribution, latency optimization.Exper
ience reducing hallucinations through data design, scoring, and retrieval tuning.Abili
ty to handle ambiguity and build in zero-
to-one, high-ownership envi
ronments.Track
record showing 50%+
time spent coding and
solving deep technical problems.
Why Join Us
Direct ownership of building a
ground-up Intelligence Platform
used across multiple products.
Solve deep engineering challenges at the intersection of
LLMs, agents, and platform architecture
.
Work with a high-caliber team in a fast-paced, high-impact environment.
Opportunity to influence systems, architecture, and long-term platform vision.