Search by job, company or skills

  • Posted 2 months ago
  • Over 200 applicants have applied

Job Description

NK Securities Research is a leading financial firm that leverages cutting-edge technology and sophisticated algorithms to trade the financial markets. Founded in 2011, we have gained invaluable experience in the field of High-Frequency Trading (HFT) across different asset classes.

Role Overview

We're looking for engineers who can take AI work beyond experiments and make it hold up in production. You'll work closely with quant researchers and infra engineers to build AI systems that actually get used improving research speed and internal tooling without slowing down the core stack. We value engineers who think about trade-offs, test what they build, and care about how things run in production.

What You'll Build

Production AI

  • Ship models that meet defined latency and reliability expectation
  • Add monitoring, rollback, and guardrails before anything goes live
  • Optimise inference across CPU/GPU environments when it matters

Integration into Real Systems

  • Plug AI into data-heavy workflows without hurting performance
  • Work within existing low-latency architecture instead of fighting it
  • Profile and remove bottlenecks rather than guessing

AI for Engineers & Researchers

  • Build tools that genuinely speed up research and development
  • Improve code understanding, review workflows, and internal knowledge retrieval
  • Keep systems auditable and predictable

LLM & Retrieval Systems

  • Implement structured RAG and embedding pipelines with validation in place
  • Create safe integration layers between models and internal systems

Performance & Standards

  • Track latency, drift, and stability — not just accuracy
  • Build observability into everything you ship
  • Help raise the bar for how AI is engineered here

What We're Looking For

Strong Python fundamentals

  • Clear thinking around system design and performance trade-offs
  • Experience deploying AI systems in production (1–5 years is typical)
  • Familiarity with transformers, embeddings, or LLM deployment

Nice to have:

  • Exposure to C++ / Rust / Go
  • Experience in distributed or performance-critical environments
  • Comfort operating with ownership and minimal hand-holding

Why This Role

  • You'll build AI systems that directly impact research and infrastructure
  • You'll work with engineers who argue about trade-offs — and care about getting them right
  • You'll have real ownership from design to deployment

What We Offer:

  • Competitive salary package.
  • A dynamic, high-performance, and collaborative work environment.
  • Strong focus on career growth and development.
  • Catered breakfast and lunch.
  • Monthly team dinners.
  • Annual international and domestic team trips.

More Info

Job Type:
Industry:
Employment Type:

Key Skills

LLM deployment

embeddings

Similar Jobs

Gurugram, India
Skills:
Ml, Tensorflow, Tuning, Jax, BigQuery, Pytorch, FastAPI, Python, Flask, Nlp, Vertex Search, Vertex AI, Gemini, vector stores, Pub Sub, embeddings, prompt design, GCP services, Generative AI models, Cloud Functions
Noida, India
Skills:
Schema Design, Python, Prompting, Ranking and re-ranking, Embeddings, Context management, Source validation, Evaluation infrastructure, Vertex AI, Retrieval quality, Metadata filtering
3-5 yrs
Gurugram, Gurugram, India
Skills:
Python, data pipelines, hallucination mitigation techniques, Mistral, MLOps tools, guardrails, RAG pipelines, fine-tuning techniques, Generative AI models, safety mechanisms, fine-tuning workflows, model monitoring, Haystack, LangChain, PEFT, instruction tuning, LLaMA, LLMs, Anthropic, prompt engineering, CI CD pipelines, LoRA, OpenAI, LlamaIndex, multimodal models
3-5 yrs
Gurugram, India, Gurugram
Skills:
Git, Pytorch, Linux, Docker, Opencv, Python, Image Processing, PEFT, NVIDIA GPUs, Transformers, image generation, Hugging Face, reinforcement learning, Diffusers
1-3 yrs
Gurugram, Gurugram, India
Skills:
Python, Sql, AI fluency, Data Quality Checks, LLMs, Data Pipelines, data infrastructure, data models