ABOUT ADGRID

AdGrid is an early-stage, stealth-mode AdTech–FinTech company based in Gurugram, building a platform at the intersection of advertising technology and financial infrastructure. We are a small, fast-moving team hiring the people who will help shape the company from the ground up.

ROLE SUMMARY

We are looking for a Machine Learning Engineer to build and ship intelligent voice and language experiences across speech, LLMs, retrieval, and memory. You will work on STT, TTS, RAG, model training, and on-device AI, taking ML systems from data and experimentation to production.

KEY RESPONSIBILITIES

- Build and optimise speech-to-text (STT) and text-to-speech (TTS) pipelines for natural, low-latency voice interactions.
- Improve speech accuracy across Indian languages, accents, and noisy real-world environments.
- Build LLM-powered features using Retrieval-Augmented Generation (RAG), embeddings, vector search, and grounded responses.
- Design and manage short-term context and long-term memory across sessions.
- Work on prompt design, model evaluation, and continuous improvement of response quality.
- Create, train, fine-tune, and evaluate models across speech, language, and retrieval use cases.
- Work across the complete ML lifecycle, including data, features, training, evaluation, and iteration.
- Run and optimise LLMs and speech models locally, on-device, or at the edge.
- Optimize models for edge deployment through quantization, compression, and efficient runtimes.
- Ship ML models to production and contribute to serving, monitoring, and retraining.
- Build reliable data and training pipelines.
- Work closely with product and engineering teams to turn AI ideas into production-ready features.





WHAT WE ARE LOOKING FOR

- 2–3 years of hands-on experience building ML and AI systems, with experience shipping toward production.
- Hands-on experience or strong project work with speech-to-text (STT) and text-to-speech (TTS) pipelines.
- Experience building LLM systems using RAG, embeddings, vector search, and model memory/context.
- Experience creating, training, fine-tuning, and evaluating ML models.
- Strong understanding of deep learning, with hands-on experience using PyTorch and good fundamentals in mathematics and statistics.
- Experience with local or edge execution of LLMs, including running and optimising models on-device or at the edge.
- Strong Python skills and solid software engineering practices, including Git and testing.
- Experience with LLMs, RAG frameworks such as LangChain/LlamaIndex, vector databases, and fine-tuning techniques such as PEFT/LoRA.
- Familiarity with model optimisation techniques such as quantisation, compression, distillation, or pruning is a plus.
- Exposure to MLOps, experiment tracking, model monitoring, Docker, CI/CD, and cloud platforms such as AWS, GCP, or Azure is a plus.
- Experience with Indian languages, multilingual or code-mixed speech/NLP, agentic systems, or real-time audio systems is an advantage.
- Good problem-solving skills and ability to work independently in a fast-moving workplace.
- Bachelor’s degree in Computer Science, Engineering, IT, or a related field.
- Available to work full-time from our Gurugram office (on-site).

HOW TO APPLY Interested candidates may send their resume to [email protected]. A GitHub profile, research paper, or link to an AI/ML project you have built is welcome.

📌 Machine Learning Engineer (Gurugram)
🏢 AdGrid
📍 Gurugram