Senior AI / ML Engineer

TekFrameworks; Hyderabad; On-site; INR 11,561,000 - 20,231,000; Full time

Job summary

TekFrameworks is seeking an AI Engineer to build core AI capabilities and intelligent applications. You will design and maintain reproducible data pipelines, work on DAPT/SFT/LoRA, and develop AI agents and model services via APIs.

The role spans data engineering, LLM training, API development, and agentic application design, collaborating with product and engineering teams to deliver scalable AI-driven solutions.

Qualifications

• 5-6 years of hands-on AI/ML/NLP engineering experience with shipped projects.
• Strong Python skills; experience with PyTorch and modern ML frameworks.
• Experience with API development (FastAPI, Flask, or similar).
• Familiarity with Linux, Git, Docker, and MLOps practices (tracking, reproducibility).
• Strong communication and collaboration skills.

Responsibilities

• Design and maintain reproducible data ingestion and transformation pipelines; ensure data quality, governance, and compliance for structured and unstructured datasets.
• Strong systems-thinking ability to design modular, scalable, and reproducible AI pipelines across data ingestion, corpus engineering, training, and deployment.
• Support domain-adaptive pre-training (DAPT) and fine-tuning (SFT, LoRA/QLoRA/PEFT) for specialized tasks; implement cost/performance optimizations.
• Strong understanding of tokenization, sequence packing, context length trade-offs, and their impact on memory and training efficiency.
• Develop AI agents capable of reasoning, planning, and executing multi-step workflows; integrate tools, APIs, and external data sources for autonomous task execution.
• API Development & Integration: Build and expose model services via RESTful APIs or gRPC for seamless integration; collaborate on architecture decisions for scalable serving stacks (vLLM/TGI/FastAPI).
• Implement end-to-end RAG pipelines: chunking, embeddings, vector indexing, retrieval, and grounded generation.
• Evaluation & Safeguards: Measure retrieval and generation quality using metrics (recall@k, MRR, nDCG, faithfulness); implement safeguards: citation enforcement, refusal policies, and uncertainty signaling.
• Documentation & Collaboration: Maintain experiment logs, configs, and reproducibility checklists; work cross-functionally to integrate AI solutions into enterprise applications.
• Lead and guide team members, providing technical direction while actively contributing as a hands‑on engineer

Skills

Python

PyTorch

APIs

MLOps

Communication

Education

Bachelor's degree in Computer Science, AI, Data Science, or related field

Tools

FastAPI

Flask

Linux

Git

Docker

Job description

We are looking for an AI Engineer to build core AI capabilities and intelligent applications. This role spans data engineering, LLM training and fine-tuning, API development, and agentic application design. You will collaborate with product, data, and engineering teams to deliver scalable, reproducible pipelines and advanced AI-driven solutions.

Key Responsibilities

We believe in ownership, creativity, and continuous learning.

• Design and maintain reproducible data ingestion and transformation pipelines; ensure data quality, governance, and compliance for structured and unstructured datasets.
• Strong systems-thinking ability to design modular, scalable, and reproducible AI pipelines across data ingestion, corpus engineering, training, and deployment.
• Support domain-adaptive pre-training (DAPT) and fine-tuning (SFT, LoRA/QLoRA/PEFT) for specialized tasks; implement cost/performance optimizations (quantization, checkpointing, hyperparameter tuning).
• Strong understanding of tokenization, sequence packing, context length trade-oAs, and their impact on memory and training eAiciency.
• Develop AI agents capable of reasoning, planning, and executing multi-step workflows; integrate tools, APIs, and external data sources for autonomous task execution.
• API Development & Integration: Build and expose model services via RESTful APIs or gRPC for seamless integration; collaborate on architecture decisions for scalable serving stacks (vLLM/TGI/FastAPI).
• Implement end-to-end RAG pipelines: chunking, embeddings, vector indexing, retrieval, and grounded generation.
• Evaluation & Safeguards: Measure retrieval and generation quality using metrics (recall@k, MRR, nDCG, faithfulness); implement safeguards: citation enforcement, refusal policies, and uncertainty signaling.
• Documentation & Collaboration: Maintain experiment logs, configs, and reproducibility checklists; work cross-functionally to integrate AI solutions into enterprise applications.
• Lead and guide team members, providing technical direction while actively contributing as a hands‑on engineer

Required Skills & Experience

• 5-6 years of hands‑on experience in AI/ML/NLP engineering with shipped projects.
• Strong Python skills; experience with PyTorch and modern ML frameworks.
• Experience with API development (FastAPI, Flask, or similar).
• Familiarity with Linux, Git, Docker, and MLOps practices (tracking, reproducibility).
• Strong communication and collaboration skills.

Preferred / Good to Have

• Exposure to multi‑GPU training stacks (DeepSpeed/FSDP) and mixed‑precision training.
• Experience serving LLMs and optimizing inference (quantization, batching, caching).
• Knowledge of agentic frameworks (LangChain, AutoGen, or similar).
• Familiarity with vector databases (e.g., Pinecone, Weaviate) and RAG evaluation tools.
• Understanding of knowledge graphs and entity‑centric evaluation.
• Cloud experience (Azure/AWS/GCP), Kubernetes, and secure deployments.

Education

Minimum: B.E./B.Tech/M.E./M.Tech/M.Sc. in Computer Science, AI/ML, Data Science, or related field (or equivalent practical experience).

50 Division Street, Suite 501,
Somerville, NJ 08876, US