Website:
blessingsofttech.com
Job details:
Principal Machine Learning Engineer (Speech AI and LLM Agents)
Company: Blessing Softtech
Location: Pune, Maharashtra, India
Workplace: Hybrid
Employment Type: Full-time
Seniority: Mid-Senior Level
About Blessing Softtech
Blessing Softtech is a bootstrapped, DPIIT-recognised AI company headquartered in Pune, building a portfolio of AI products used across finance, healthcare, mobility and real estate. The engineering team is small, senior and ships to production every week.
The Role
You will be our first principal-level ML hire. You set the technical direction for realtime speech and LLM agent systems, write production code every week, and hire the ML team that follows you. You report directly to the Founder and CEO.
The bar is simple: systems that hold a natural spoken conversation at under 300 ms p50 voice-to-voice latency, in dozens of languages, with many speakers and many agents at once, measured and improved every day.
What You Will Do
- Own the full realtime speech loop: streaming ASR, LLM inference and streaming TTS, with hard p50 and p95 latency budgets for every hop.
- Design, train and ship turn-taking, endpointing, barge-in and audio emotion models where off-the-shelf systems fall short.
- Build agent systems that compile a natural-language brief into a tested production agent: tool use, structured state graphs, automated simulation, scoring and self-repair.
- Architect multi-party, multilingual realtime audio: per-speaker streams, live speech translation, and floor control between humans and agents.
- Implement agent-to-agent communication on open protocols (A2A, MCP) with verified identity, scoped permissions and loop protection.
- Build the evaluation and data flywheel: golden sets per language, LLM-as-judge calibrated against human labels, regression gates in CI, consent-aware labelling and PII redaction.
- Fine-tune, distil, quantise and serve open-weight LLM and speech models, on cloud GPUs and on edge hardware that runs with no network.
- Automate MLOps end to end: training pipelines, model registry, canary releases, auto rollback, and cost and latency guardrails per tenant.
- Hire, mentor and lead a team of 3 to 8 ML engineers.
What You Bring
- 7+ years building ML systems, with 3+ years shipping LLM or speech systems to real users at scale.
- A latency-critical streaming system you shipped, and the numbers to prove it: p50, p95, cost per unit, failure modes.
- Expert Python and PyTorch. Strong command of transformer internals: attention variants, KV cache, decoding and sampling strategies.
- Hands-on speech experience: Conformer and Whisper-class ASR, neural codec or flow-matching TTS, VAD, audio front ends, Opus and telephony sample rates.
- LLM serving and optimisation: vLLM, TensorRT-LLM or equivalent; continuous batching, speculative decoding, INT8 and INT4 quantisation (AWQ, GPTQ).
- Training and alignment: SFT, LoRA and QLoRA, DPO or other preference optimisation, distillation; evals that predict production outcomes.
- Production infrastructure: Kubernetes, AWS, GPU scheduling, tracing and observability.
- You write clearly, turn ambiguity into a spec, and ship.
Strong Plus
- WebRTC, SIP and RTP, or telephony at scale.
- Multilingual and Indic speech, including code-switched speech.
- Multi-agent systems in production; A2A or MCP.
- On-device inference: llama.cpp, ONNX Runtime, TensorRT on Jetson, Core ML.
- Audio watermarking, voice anti-spoofing or deepfake detection.
- Publications at Interspeech, ICASSP, ACL, EMNLP or NeurIPS, or widely used open-source work.
- Has hired and led ML engineers.
Stack
Python, PyTorch, Hugging Face, vLLM, Triton Inference Server, Kubernetes on AWS EKS, Amazon Bedrock, Langfuse, Redis, WebRTC.
How We Hire
- 30 min intro screen
- 45 min conversation with the Founder
- 60 min technical deep dive: realtime speech and LLM agents
- 90 min live system design
- Paid take-home (4 to 6 hours) or an ML depth session, your choice
- References
Decision within 10 working days of your first call.
What You Get
- First principal ML seat, with real ownership of architecture and hiring.
- Direct line to the Founder; no layers.
- Hard, unsolved problems in realtime speech and multi-agent systems.
- Freedom to publish and open-source non-core work.
- Top-of-market compensation for the role.
How to Apply
Include a link to the system, paper or repo you are proudest of, and one latency or quality number you personally moved.
Blessing Softtech is an equal opportunity employer. We hire on skill and evidence of work, regardless of background, gender, age, religion, caste or disability.
Click on Apply to know more.
